![research paper about modular distance learning U.S. flag](https://www.ncbi.nlm.nih.gov/coreutils/uswds/img/favicons/favicon-57.png)
An official website of the United States government
The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.
The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.
- Publications
- Account settings
Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .
- Advanced Search
- Journal List
- Behav Sci (Basel)
![research paper about modular distance learning Logo of behavsci](https://www.ncbi.nlm.nih.gov/corehtml/pmc/pmcgifs/logo-behavsci.png)
![](http://sektorel.online/777/templates/cheerup/res/banner1.gif)
Assessing Cognitive Factors of Modular Distance Learning of K-12 Students Amidst the COVID-19 Pandemic towards Academic Achievements and Satisfaction
Yung-tsan jou.
1 Department of Industrial and Systems Engineering, Chung Yuan Christian University, Taoyuan 320, Taiwan; wt.ude.ucyc@uojty (Y.-T.J.); moc.oohay@enimrahcrolfas (C.S.S.)
Klint Allen Mariñas
2 School of Industrial Engineering and Engineering Management, Mapua University, Manila 1002, Philippines
3 Department of Industrial Engineering, Occidental Mindoro State College, San Jose 5100, Philippines
Charmine Sheena Saflor
Associated data.
Not applicable.
The COVID-19 pandemic brought extraordinary challenges to K-12 students in using modular distance learning. According to Transactional Distance Theory (TDT), which is defined as understanding the effects of distance learning in the cognitive domain, the current study constructs a theoretical framework to measure student satisfaction and Bloom’s Taxonomy Theory (BTT) to measure students’ academic achievements. This study aims to evaluate and identify the possible cognitive capacity influencing K-12 students’ academic achievements and satisfaction with modular distance learning during this new phenomenon. A survey questionnaire was completed through an online form by 252 K-12 students from the different institutions of Occidental Mindoro. Using Structural Equation Modeling (SEM), the researcher analyses the relationship between the dependent and independent variables. The model used in this research illustrates cognitive factors associated with adopting modular distance learning based on students’ academic achievements and satisfaction. The study revealed that students’ background, experience, behavior, and instructor interaction positively affected their satisfaction. While the effects of the students’ performance, understanding, and perceived effectiveness were wholly aligned with their academic achievements. The findings of the model with solid support of the integrative association between TDT and BTT theories could guide decision-makers in institutions to implement, evaluate, and utilize modular distance learning in their education systems.
1. Introduction
The 2019 coronavirus is the latest infectious disease to develop rapidly worldwide [ 1 ], affecting economic stability, global health, and education. Most countries have suspended thee-to-face classes in order to curb the spread of the virus and reduce infections [ 2 ]. One of the sectors impacted has been education, resulting in the suspension of face-to-face classes to avoid spreading the virus. The Department of Education (DepEd) has introduced modular distance learning for K-12 students to ensure continuity of learning during the COVID-19 pandemic. According to Malipot (2020), modular learning is one of the most popular sorts of distance learning alternatives to traditional face-to-face learning [ 3 ]. As per DepEd’s Learner Enrolment and Survey Forms, 7.2 million enrollees preferred “modular” remote learning, TV and radio-based practice, and other modalities, while two million enrollees preferred online learning. It is a method of learning that is currently being used based on the preferred distance learning mode of the students and parents through the survey conducted by the Department of Education (DepEd); this learning method is mainly done through the use of printed and digital modules [ 4 ]. It also concerns first-year students in rural areas; the place net is no longer available for online learning. Supporting the findings of Ambayon (2020), modular teaching within the teach-learn method is more practical than traditional educational methods because students learn at their own pace during this modular approach. This educational platform allows K-12 students to interact in self-paced textual matter or digital copy modules. With these COVID-19 outbreaks, some issues concerned students’ academic, and the factors associated with students’ psychological status during the COVID-19 lockdown [ 5 ].
Additionally, this new learning platform, modular distance learning, seems to have impacted students’ ability to discover and challenged their learning skills. Scholars have also paid close attention to learner satisfaction and academic achievement when it involves distance learning studies and have used a spread of theoretical frameworks to assess learner satisfaction and educational outcomes [ 6 , 7 ]. Because this study aimed to boost academic achievement and satisfaction in K-12 students, the researcher thoroughly applied transactional distance theory (TDT) to understand the consequences of distance in relationships in education. The TDT was utilized since it has the capability to establish the psychological and communication factors between the learners and the instructors in distance education that could eventually help researchers in identifying the variables that might affect students’ academic achievement and satisfaction [ 8 ]. In this view, distance learning is primarily determined by the number of dialogues between student and teacher and the degree of structuring of the course design. It contributes to the core objective of the degree to boost students’ modular learning experiences in terms of satisfaction. On the other hand, Bloom’s Taxonomy Theory (BTT) was applied to investigate the students’ academic achievements through modular distance learning [ 6 ]. Bloom’s theory was employed in addition to TDT during this study to enhance students’ modular educational experiences. Moreover, TDT was utilized to check students’ modular learning experiences in conjuction with enhacing students’ achievements.
This study aimed to detect the impact of modular distance learning on K-12 students during the COVID-19 pandemic and assess the cognitive factors affecting academic achievement and student satisfaction. Despite the challenging status of the COVID-19 outbreak, the researcher anticipated a relevant result of modular distance learning and pedagogical changes in students, including the cognitive factors identified during this paper as latent variables as possible predictors for the utilization of K-12 student academic achievements and satisfaction.
1.1. Theoretical Research Framework
This study used TDT to assess student satisfaction and Bloom’s theory to quantify academic achievement. It aimed to assess the impact of modular distance learning on academic achievement and student satisfaction among K-12 students. The Transactional Distance Theory (TDT) was selected for this study since it refers to student-instructor distance learning. TDT Moore (1993) states that distance education is “the universe of teacher-learner connections when learners and teachers are separated by place and time.” Moore’s (1990) concept of ”Transactional Distance” adopts the distance that occurs in all linkages in education, according to TDT Moore (1993). Transactional distance theory is theoretically critical because it states that the most important distance is transactional in distance education, rather than geographical or temporal [ 9 , 10 ]. According to Garrison (2000), transactional distance theory is essential in directing the complicated experience of a cognitive process such as distance teaching and learning. TDT evaluates the role of each of these factors (student perception, discourse, and class organization), which can help with student satisfaction research [ 11 ]. Bloom’s Taxonomy is a theoretical framework for learning created by Benjamin Bloom that distinguishes three learning domains: Cognitive domain skills center on knowledge, comprehension, and critical thinking on a particular subject. Bloom recognized three components of educational activities: cognitive knowledge (or mental abilities), affective attitude (or emotions), and psychomotor skills (or physical skills), all of which can be used to assess K-12 students’ academic achievement. According to Jung (2001), “Transactional distance theory provides a significant conceptual framework for defining and comprehending distance education in general and a source of research hypotheses in particular,” shown in Figure 1 [ 12 ].
![Click on image to zoom An external file that holds a picture, illustration, etc.
Object name is behavsci-12-00200-g001.jpg](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9312230/bin/behavsci-12-00200-g001.jpg)
Theoretical Research Framework.
1.2. Hypothesis Developments and Literature Review
This section will discuss the study hypothesis and relate each hypothesis to its related studies from the literature.
There is a significant relationship between students’ background and students’ behavior .
The teacher’s guidance is essential for students’ preparedness and readiness to adapt to a new educational environment. Most students opt for the Department of Education’s “modular” distance learning options [ 3 ]. Analyzing students’ study time is critical for behavioral engagement because it establishes if academic performance is the product of student choice or historical factors [ 13 ].
There is a significant relationship between students’ background and students’ experience .
Modules provide goals, experiences, and educational activities that assist students in gaining self-sufficiency at their speed. It also boosts brain activity, encourages motivation, consolidates self-satisfaction, and enables students to remember what they have learned [ 14 ]. Despite its success, many families face difficulties due to their parents’ lack of skills and time [ 15 ].
There is a significant relationship between students’ behavior and students’ instructor interaction .
Students’ capacity to answer problems reflects their overall information awareness [ 5 ]. Learning outcomes can either cause or result in students and instructors behavior. Students’ reading issues are due to the success of online courses [ 16 ].
There is a significant relationship between students’ experience and students’ instructor interaction .
The words “student experience” relate to classroom participation. They establish a connection between students and their school, teachers, classmates, curriculum, and teaching methods [ 17 ]. The three types of student engagement are behavioral, emotional, and cognitive. Behavioral engagement refers to a student’s enthusiasm for academic and extracurricular activities. On the other hand, emotional participation is linked to how children react to their peers, teachers, and school. Motivational engagement refers to a learner’s desire to learn new abilities [ 18 ].
There is a significant relationship between students’ behavior and students’ understanding .
Individualized learning connections, outstanding training, and learning culture are all priorities at the Institute [ 19 , 20 ]. The modular technique of online learning offers additional flexibility. The use of modules allows students to investigate alternatives to the professor’s session [ 21 ].
There is a significant relationship between students’ experience and students’ performance .
Student conduct is also vital in academic accomplishment since it may affect a student’s capacity to study as well as the learning environment for other students. Students are self-assured because they understand what is expected [ 22 ]. They are more aware of their actions and take greater responsibility for their learning.
There is a significant relationship between students’ instructor interaction and students’ understanding .
Modular learning benefits students by enabling them to absorb and study material independently and on different courses. Students are more likely to give favorable reviews to courses and instructors if they believe their professors communicated effectively and facilitated or supported their learning [ 23 ].
There is a significant relationship between students’ instructor interaction and students’ performance.
Students are more engaged and active in their studies when they feel in command and protected in the classroom. Teachers play an essential role in influencing student academic motivation, school commitment, and disengagement. In studies on K-12 education, teacher-student relationships have been identified [ 24 ]. Positive teacher-student connections improve both teacher attitudes and academic performance.
There is a significant relationship between students’ understanding and students’ satisfaction .
Instructors must create well-structured courses, regularly present in their classes, and encourage student participation. When learning objectives are completed, students better understand the course’s success and learning expectations. “Constructing meaning from verbal, written, and graphic signals by interpreting, exemplifying, classifying, summarizing, inferring, comparing, and explaining” is how understanding is characterized [ 25 ].
There is a significant relationship between students’ performance and student’s academic achievement .
Academic emotions are linked to students’ performance, academic success, personality, and classroom background [ 26 ]. Understanding the elements that may influence student performance has long been a goal for educational institutions, students, and teachers.
There is a significant relationship between students’ understanding and students’ academic achievement .
Modular education views each student as an individual with distinct abilities and interests. To provide an excellent education, a teacher must adapt and individualize the educational curriculum for each student. Individual learning may aid in developing a variety of exceptional and self-reliant attributes [ 27 ]. Academic achievement is the current level of learning in the Philippines [ 28 ].
There is a significant relationship between students’ performance and students’ satisfaction .
Academic success is defined as a student’s intellectual development, including formative and summative assessment data, coursework, teacher observations, student interaction, and time on a task [ 29 ]. Students were happier with course technology, the promptness with which content was shared with the teacher, and their overall wellbeing [ 30 ].
There is a significant relationship between students’ academic achievement and students’ perceived effectiveness .
Student satisfaction is a short-term mindset based on assessing students’ educational experiences [ 29 ]. The link between student satisfaction and academic achievement is crucial in today’s higher education: we discovered that student satisfaction with course technical components was linked to a higher relative performance level [ 31 ].
There is a significant relationship between students’ satisfaction and students’ perceived effectiveness.
There is a strong link between student satisfaction and their overall perception of learning. A satisfied student is a direct effect of a positive learning experience. Perceived learning results had a favorable impact on student satisfaction in the classroom [ 32 ].
2. Materials and Methods
2.1. participants.
The principal area under study was San Jose, Occidental Mindoro, although other locations were also accepted. The survey took place between February and March 2022, with the target population of K-12 students in Junior and Senior High Schools from grades 7 to 12, aged 12 to 20, who are now implementing the Modular Approach in their studies during the COVID-19 pandemic. A 45-item questionnaire was created and circulated online to collect the information. A total of 300 online surveys was sent out and 252 online forms were received, a total of 84% response rate [ 33 ]. According to several experts, the sample size for Structural Equation Modeling (SEM) should be between 200 and 500 [ 34 ].
2.2. Questionnaire
The theoretical framework developed a self-administered test. The researcher created the questionnaire to examine and discover the probable cognitive capacity influencing K-12 students’ academic achievement in different parts of Occidental Mindoro during this pandemic as well as their satisfaction with modular distance learning. The questionnaire was designed through Google drive as people’s interactions are limited due to the effect of the COVID-19 pandemic. The questionnaire’s link was sent via email, Facebook, and other popular social media platforms.
The respondents had to complete two sections of the questionnaire. The first is their demographic information, including their age, gender, and grade level. The second is about their perceptions of modular learning. The questionnaire is divided into 12 variables: (1) Student’s Background, (2) Student’s Experience, (3) Student’s Behavior, (4) Student’s Instructor Interaction, (5) Student’s Performance, (6) Student’s Understanding, (7) Student’s Satisfaction, (8) Student’s Academic Achievement, and (9) Student’s Perceived Effectiveness. A 5-point Likert scale was used to assess all latent components contained in the SEM shown in Table 1 .
The construct and measurement items.
Construct | Items | Measures | Supporting Reference |
---|---|---|---|
Students’ Background (SB) | SB1 | Are you having difficulty with Modular Distance Learning. | Pe Dangle, Y.R. (2020) [ ] |
SB2 | I prefer Modular Distance Learning Rather than traditional face-to-face training, I prefer the Modular Distance Learning Approach. | Aksan, J.A. (2021) [ ] | |
SB3 | Modular learning aids students in increasing their productivity in education and learning while promoting flexibility in terms of content, time, and space. | Shuja, A. et al. (2019) [ ] | |
SB4 | I have a lot of time to answer the activities with a modular teaching technique. | Aksan, J.A. (2021) [ ] | |
SB5 | I acquire the same amount of learning from using the module as I do from learning in a face-to-face or classroom situation. | Natividad, E. (2021) [ ] | |
Students’ Behavior (SBE) | SBE1 | I feel confident in studying and performing well in the modular class. | Delfino, A.P. (2019) [ ] |
SBE2 | I employed rehearsing techniques like reviewing my notes over and over again. | Lowerison et al. (2006) [ ] | |
SBE3 | I can recall my understanding from the past and help me to understand words. | Santillan, S.C. et al. (2021) [ ] | |
SBE4 | I retain a critical mindset throughout my studies, considering before accepting or rejecting. | Bordeos (2021) [ ] | |
SBE5 | Usually I plan my weekly module work in advance. | Karababa et al. (2010) [ ] | |
Students’ Experience (SE) | SE1 | The way the module materials were presented helped to maintain my interest. | Allen et al. (2020) [ ] |
SE2 | I do not experience any problems during modular distance learning | Amir al (2020) [ ] | |
SE3 | The instructions for completing the assessed tasks were simple to understand. | Santillan et al. (2021) [ ] | |
SE4 | During distance learning, I am not stressed. | Amir et al. (2020) [ ] | |
SE5 | The study workload on this module fitted with my personal circumstances. | Allen et al. (2020) [ ] | |
Students’ Instructor Interaction (SI) | SI1 | The instructor updated me on my progress in the course regularly. | Gray & DiLoreto (2020) [ ] |
SI2 | On this subject, I was satisfied with my teacher’s assistance. | Allen et al. (2020) [ ] | |
SI3 | I kept in touch with the course’s instructor regularly. | Gray & DiLoreto (2020) [ ] | |
SI4 | The instructor was concerned about my performance in this class. | Gray & DiLoreto (2020) [ ] | |
SI5 | My teacher feedback on assessed tasks helped me prepare for the next assessment. | Allen et al. (2020) [ ] | |
Students’ Understanding (SAU) | SAU1 | Modular Distance Learning allows me to take my time to understand my school works. | Abuhassna et al. (2020) [ ] |
SAU2 | The distance learning program met my expectations in terms of quality. | Woolf et al. (2020) [ ] | |
SAU3 | Modular Distance Learning helps me to improve my understanding and skills and also helps to gather new knowledge. | Bordeos (2021) [ ] | |
SAU4 | Modular Distance Learning is a helpful tool to get so focused on activities in my classes. | Abuhassna et al. (2020) [ ] | |
SAU5 | Modular Distance Learning motivates me to study more about the course objectives. | Abuhassna et al. (2020) [ ] | |
Students’ Performance (SP) | SP1 | I can effectively manage my study time and complete assignments on schedule. | Richardson and Swan (2003) [ ] |
SP2 | When completing projects or participating in class discussions, combine ideas or concepts from several courses. | Delfino, A.P. (2019) [ ] | |
SP3 | I employed elaboration techniques like summarizing the material and relating it to previous knowledge. | Lowerison et al. (2006) [ ] | |
SP4 | In my studies, I am self-disciplined and find it easy to schedule reading and homework time. | Richardson and Swan (2003) [ ] | |
SP5 | I was confident in my capacity to learn and do well in class. | Delfino, A.P. (2019) [ ] | |
Student’s Academic Achievement (SAA) | SAA1 | I have more opportunities to reflect on what I’ve learned in modular classes. | Dziuban et al. (2015) [ ] |
SAA2 | I am committed to completing my homework (readings, assignments) on time and engaging fully in class discussions. | Mt. San Antonio College (2012) [ ] | |
SAA3 | My modular learning experience has increased my opportunity to access and use information. | Dziuban et al. (2015) [ ] | |
SAA4 | I employed assessment, evaluation, and criticizing procedures for assessing, evaluating, and critiquing the material. | Lowerison et al. (2006) [ ] | |
SAA5 | I am skilled at juggling many responsibilities while working under time constraints. | Estelami (2013) [ ] | |
Students’ Satisfaction (SS) | SS1 | I am always interested in learning about new things. | Abuhassna et al. (2020) [ ] |
SS2 | I study more efficiently with distance learning. | Amir (2020) [ ] | |
SS3 | Modular learning suits me better than face-to-face classes. | Abuhassna et al. (2020) [ ] | |
SS4 | I prefer distance learning to classroom learning. | Amir et al. (2020) [ ] | |
SS5 | Overall, I am pleased with the module’s quality. | Santillan, S.C. et al. (2021) [ ] | |
Students’ Perceived Effectiveness (SPE) | SPE1 | I made use of learning possibilities and resources in this modular distance learning. | Lowerison et al. (2006) [ ] |
SPE2 | I would recommend modular distance learning study to other students. | Abuhassna et al. (2020) [ ] | |
SPE3 | These classes also challenge me to conduct more independent research and not rely on a single source of information. | Mt. San Antonio College (2012) [ ] | |
SPE4 | Overall, this modular distance learning has been a good platform for studying during the pandemic. | Lowerison et al., 2006) [ ] | |
SPE5 | Overall, I am satisfied with this modular distance learning course. | Aman (2009) [ ] |
2.3. Structural Equation Modeling (SEM)
All the variables have been adapted from a variety of research in the literature. The observable factors were scored on a Likert scale of 1–5, with one indicating “strongly disagree” and five indicating “strongly agree”, and the data were analyzed using AMOS software. Theoretical model data were confirmed by Structural Equation Modeling (SEM). SEM is more suitable for testing the hypothesis than other methods [ 53 ]. There are many fit indices in the literature, of which the most commonly used are: CMIN/DF, Comparative Fit Index (CFI), AGFI, GFI, and Root Mean Square Error (RMSEA). Table 2 demonstrates the Good Fit Values and Acceptable Fit Values of the fit indices, respectively. AGFI and GFI are based on residuals; when sample size increases, the value of the AGFI also increase. It takes a value between 0 and 1. The fit is good if the value is more significant than 0.80. GFI is a model index that spans from 0 to 1, with values above 0.80 deemed acceptable. An RMSEA of 0.08 or less suggests a good fit [ 54 ], and a value of 0.05 to 0.08 indicates an adequate fit [ 55 ].
Acceptable Fit Values.
Fit Indices | Acceptable Range | Reference |
---|---|---|
CMIN/DF | <3.00 | Norberg et al., 2007 [ ]; Li et al., 2013 [ ] |
GFI | ≥0.80 | Doloi et al., 2012 [ ] |
CFI | >0.70 | Norberg et al., 2007 [ ]; Chen et al., 2012 [ ] |
RMSEA | ≤0.08 | Doloi et al., 2012 [ ] |
AGFI | >0.08 | Jaccard and Wan (1996) [ ] |
TLI | >0.08 | Jafari et al., 2021 [ ] |
IFI | >0.08 | Lee et al., 2015 [ ] |
3. Results and Discussion
Figure 2 demonstrates the initial SEM for the cognitive factors of Modular Distance learning towards academic achievements and satisfaction of K-12 students during the COVID-19 pandemic. According to the figure below, three hypotheses were not significant: Students’ Behavior to Students’ Instructor Interaction (Hypothesis 3), Students’ Understanding of Students’ Academic Achievement (Hypothesis 11), and Students’ Performance to Students’ Satisfaction (Hypothesis 12). Therefore, a revised SEM was derived by removing this hypothesis in Figure 3 . We modified some indices to enhance the model fit based on previous studies using the SEM approach [ 47 ]. Figure 3 demonstrates the final SEM for evaluating cognitive factors affecting academic achievements and satisfaction and the perceived effectiveness of K-12 students’ response to Modular Learning during COVID-19, shown in Table 3 . Moreover, Table 4 demonstrates the descriptive statistical results of each indicator.
![Click on image to zoom An external file that holds a picture, illustration, etc.
Object name is behavsci-12-00200-g002.jpg](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9312230/bin/behavsci-12-00200-g002.jpg)
Initial SEM with indicators for evaluating the cognitive factors of modular distance learning towards academic achievements and satisfaction of K-12 students during COVID-19 pandemic.
![Click on image to zoom An external file that holds a picture, illustration, etc.
Object name is behavsci-12-00200-g003.jpg](https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9312230/bin/behavsci-12-00200-g003.jpg)
Revised SEM with indicators for evaluating the cognitive factors of modular distance learning towards academic achievements and satisfaction of K-12 students during the COVID-19 pandemic.
Summary of the Results.
Hypothesis | -Value | Interpretation | |
---|---|---|---|
H1 | There is a significant relationship between Students’ Background and Students’ Behavior | 0.001 | Significant |
H2 | There is a significant relationship between Students’ Background and Students’ Experiences. | 0.001 | Significant |
H3 | There is a significant relationship between Students’ Behavior and Students’ instructor Interaction. | 0.155 | Not Significant |
H4 | There is a significant relationship between Students’ experience and Students—Interaction | 0.020 | Significant |
H5 | There is a significant relationship between Students’ Behavior and Students’ Understanding | 0.212 | Not Significant |
H6 | There is a significant relationship between Students’ experience and Students’ Performance | 0.001 | Significant |
H7 | There is a significant relationship between Students’ instructor Interaction and Students’ Understanding | 0.008 | Significant |
H8 | There is a significant relationship between Students’ Instructor—Interaction and students’ Performance | 0.018 | Significant |
H9 | There is a significant relationship between students’ Understanding and Students’ Satisfaction | 0.001 | Significant |
H10 | There is a significant relationship between students’ Performance and Students’ Academic Achievement | 0.001 | Significant |
H11 | There is a significant relationship between students’ understanding and Students’ Academic Achievement | 0.001 | Significant |
H12 | There is a significant relationship between students’ Performance and Students Satisfaction | 0.602 | Not Significant |
H13 | There is a significant relationship between Students’ Academic Achievement and students’ Perceived Effectiveness | 0.001 | Significant |
H14 | There is a significant relationship between students’ Satisfaction and Students’ Perceived Effectiveness | 0.001 | Significant |
Descriptive statistic results.
Factor | Item | Mean | SD | Factor Loading | |
---|---|---|---|---|---|
Initial Model | Final Model | ||||
Students’ Background | SB1 | 3.437 | 0.9147 | 0.052 | - |
SB2 | 3.000 | 1.1707 | 0.480 | 0.562 | |
SB3 | 3.663 | 0.8042 | 0.551 | 0.551 | |
SB4 | 3.742 | 0.8797 | 0.381 | - | |
SB5 | 3.024 | 1.0746 | 0.557 | 0.629 | |
Students’ Behavior | SBE1 | 3.667 | 0.8468 | 0.551 | - |
SBE2 | 3.829 | 0.8075 | 0.463 | - | |
SBE3 | 3.873 | 0.7305 | 0.507 | - | |
SBE4 | 3.833 | 0.8157 | 0.437 | - | |
SBE 5 | 3.849 | 0.8188 | 0.585 | - | |
Students’ Experience | SE1 | 3.853 | 0.8458 | 0.669 | 0.686 |
SE2 | 2.825 | 1.0643 | 0.572 | 0.525 | |
SE3 | 3.591 | 0.9035 | 0.630 | 0.634 | |
SE4 | 2.758 | 1.1260 | 0.639 | 0.611 | |
SE5 | 3.615 | 0.8693 | 0.552 | 0.551 | |
Students’ Instructor Interaction | SI1 | 3.754 | 0.7698 | 0.689 | - |
SI2 | 3.817 | 0.9095 | 0.655 | 0.541 | |
SI3 | 3.730 | 0.9269 | 0.731 | 0.645 | |
SI4 | 3.929 | 0.7487 | 0.685 | 0.568 | |
SI5 | 3.909 | 0.7805 | 0.669 | 0.597 | |
Students’ Understanding | SAU1 | 4.028 | 0.8152 | 0.647 | 0.652 |
SAU2 | 3.464 | 0.8390 | 0.691 | 0.704 | |
SAU3 | 3.873 | 0.8325 | 0.658 | 0.620 | |
SAU4 | 3.750 | 0.8775 | 0.731 | 0.741 | |
SAU5 | 3.794 | 0.8591 | 0.740 | 0.717 | |
Students’ Performance | SP1 | 3.762 | 0.8602 | 0.640 | - |
SP2 | 3.881 | 0.7995 | 0.708 | 0.655 | |
SP3 | 3.778 | 0.8781 | 0.582 | 0.606 | |
SP4 | 3.905 | 0.7927 | 0.647 | 0.585 | |
SP5 | 3.976 | 0.8275 | 0.630 | 0.673 | |
Student’s Academic Achievement | SAA1 | 3.885 | 0.7976 | 0.696 | 0.713 |
SAA2 | 3.929 | 0.7749 | 0.653 | 0.658 | |
SAA3 | 3.762 | 0.8319 | 0.632 | 0.615 | |
SAA4 | 3.837 | 0.7790 | 0.612 | 0.597 | |
SAA5 | 3.694 | 0.8959 | 0.559 | - | |
Students’ Satisfaction | SS1 | 4.087 | 0.7835 | 0.189 | - |
SS2 | 3.361 | 0.9943 | 0.657 | 0.669 | |
SS3 | 2.960 | 1.1390 | 0.779 | 0.659 | |
SS4 | 2.889 | 1.1516 | 0.759 | 0.677 | |
SS5 | 3.377 | 1.0918 | 0.802 | 0.803 | |
Students’ Perceived Effectiveness | SPE1 | 3.829 | 0.7875 | 0.580 | 0.558 |
SPE2 | 3.405 | 1.0153 | 0.730 | 0.750 | |
PE3 | 3.790 | 0.8178 | 0.490 | - | |
PE4 | 3.813 | 0.9367 | 0.614 | - | |
PE5 | 3.492 | 1.0000 | 0.696 | 0.690 |
The current study was improved by Moore’s transactional distance theory (TDT) and Bloom’s taxonomy theory (BTT) to evaluate cognitive factors affecting academic achievements and satisfaction and the perceived effectiveness of K-12 students’ response toward modular learning during COVID-19. SEM was utilized to analyze the correlation between Student Background (SB), Student Experience (SE), Student Behavior (SBE), Student Instructor Interaction (SI), Student Performance (SP), Student Understanding (SAU), Student Satisfaction (SS), Student’s Academic achievement (SAA), and Student’s Perceived effectiveness (SPE). A total of 252 data samples were acquired through an online questionnaire.
According to the findings of the SEM, the students’ background in modular learning had a favorable and significant direct effect on SE (β: 0.848, p = 0.009). K-12 students should have a background and knowledge in modular systems to better experience this new education platform. Putting the students through such an experience would support them in overcoming all difficulties that arise due to the limitations of the modular platforms. Furthermore, SEM revealed that SE had a significant adverse impact on SI (β: 0.843, p = 0.009). The study shows that students who had previous experience with modular education had more positive perceptions of modular platforms. Additionally, students’ experience with modular distance learning offers various benefits to them and their instructors to enhance students’ learning experiences, particularly for isolated learners.
Regarding the Students’ Interaction—Instructor, it positively impacts SAU (β: 0.873, p = 0.007). Communication helps students experience positive emotions such as comfort, satisfaction, and excitement, which aim to enhance their understanding and help them attain their educational goals [ 62 ]. The results revealed that SP substantially impacted SI (β: 0.765; p = 0.005). A student becomes more academically motivated and engaged by creating and maintaining strong teacher-student connections, which leads to successful academic performance.
Regarding the Students’ Understanding Response, the results revealed that SAA (β: 0.307; p = 0.052) and SS (β: 0.699; p = 0.008) had a substantial impact on SAU. Modular teaching is concerned with each student as an individual and with their specific capability and interest to assist each K-12 student in learning and provide quality education by allowing individuality to each learner. According to the Department of Education, academic achievement is the new level for student learning [ 63 ]. Meanwhile, SAA was significantly affected by the Students’ Performance Response (β: 0.754; p = 0.014). It implies that a positive performance can give positive results in student’s academic achievement, and that a negative performance can also give negative results [ 64 ]. Pekrun et al. (2010) discovered that students’ academic emotions are linked to their performance, academic achievement, personality, and classroom circumstances [ 26 ].
Results showed that students’ academic achievement significantly positively affects SPE (β: 0.237; p = 0.024). Prior knowledge has had an indirect effect on academic accomplishment. It influences the amount and type of current learning system where students must obtain a high degree of mastery [ 65 ]. According to the student’s opinion, modular distance learning is an alternative solution for providing adequate education for all learners and at all levels in the current scenario under the new education policy [ 66 ]. However, the SEM revealed that SS significantly affected SPE (β: 0.868; p = 0.009). Students’ perceptions of learning and satisfaction, when combined, can provide a better knowledge of learning achievement [ 44 ]. Students’ perceptions of learning outcomes are an excellent predictor of student satisfaction.
Since p -values and the indicators in Students’ Behavior are below 0.5, therefore two paths connecting SBE to students’ interaction—instructor (0.155) and students’ understanding (0.212) are not significant; thus, the latent variable Students’ Behavior has no effect on the latent variable Students’ Satisfaction and academic achievement as well as perceived effectiveness on modular distance learning of K12 students. This result is supported by Samsen-Bronsveld et al. (2022), who revealed that the environment has no direct influence on the student’s satisfaction, behavior engagement, and motivation to study [ 67 ]. On the other hand, the results also showed no significant relationship between Students’ Performance and Students’ Satisfaction (0.602) because the correlation p -values are greater than 0.5. Interestingly, this result opposed the other related studies. According to Bossman & Agyei (2022), satisfaction significantly affects performance or learning outcomes [ 68 ]. In addition, it was discovered that the main drivers of the students’ performance are the students’ satisfaction [ 64 , 69 ].
The result of the study implies that the students’ satisfaction serves as the mediator between the students’ performance and the student-instructor interaction in modular distance learning for K-12 students [ 70 ].
Table 5 The reliabilities of the scales used, i.e., Cronbach’s alphas, ranged from 0.568 to 0.745, which were in line with those found in other studies [ 71 ]. As presented in Table 6 , the IFI, TLI, and CFI values were greater than the suggested cutoff of 0.80, indicating that the specified model’s hypothesized construct accurately represented the observed data. In addition, the GFI and AGFI values were 0.828 and 0.801, respectively, indicating that the model was also good. The RMSEA value was 0.074, lower than the recommended value. Finally, the direct, indirect, and total effects are presented in Table 7 .
Construct Validity Model.
Factor | Number of Items | Cronbach’s α |
---|---|---|
Students’ Background | 3 | 0.598 |
Students’ Behavior | 5 | 0.682 |
Students’ Experience | 5 | 0.761 |
Students’ Instructor Interaction | 5 | 0.817 |
Students’ Understanding | 5 | 0.825 |
Students’ Performance | 5 | 0.768 |
Students’ Academic Achievement | 5 | 0.770 |
Students’ Satisfaction | 5 | 0.777 |
Students’ Perceived Effectiveness | 5 | 0.772 |
Total | 0.752 |
Goodness of Fit Measures of SEM | Parameter Estimates | Minimum Cut-Off | Interpretation |
---|---|---|---|
CMIN/DF | 2.375 | <3.0 | Acceptable |
Comparative Fit Index (CFI) | 0.830 | >0.8 | Acceptable |
Incremental Fit Index (IFI) | 0.832 | >0.8 | Acceptable |
Tucker Lewis Index (TLI) | 0.812 | >0.8 | Acceptable |
Goodness of Fit Index (GFI) | 0.812 | >0.8 | Acceptable |
Adjusted Goodness of Fit Index (AGFI) | 0.803 | >0.8 | Acceptable |
Root Mean Square Error (RMSEA) | 0.074 | <0.08 | Acceptable |
Direct effect, indirect effect, and total effect.
No. | Variable | Direct Effects | -Value | Indirect Effects | -Value | Total Effects | -Value |
---|---|---|---|---|---|---|---|
1 | SB–SE | 0.848 | 0.009 | - | - | 0.848 | 0.009 |
2 | SB–SI | - | - | 0.715 | 0.006 | 0.715 | 0.006 |
3 | SB–SAU | - | - | 0.624 | 0.006 | 0.624 | 0.006 |
4 | SB–SP | - | - | 0.547 | 0.004 | 0.547 | 0.004 |
5 | SB–SAA | - | - | 0.604 | 0.006 | 0.604 | 0.006 |
6 | SB–SS | - | - | 0.436 | 0.006 | 0.436 | 0.006 |
7 | SB–SPE | - | - | 0.522 | 0.007 | 0.522 | 0.006 |
8 | SE–SI | 0.843 | 0.009 | - | - | 0.843 | 0.009 |
9 | SE–SAU | - | - | 0.736 | 0.010 | 0.736 | 0.010 |
10 | SE–SP | - | - | 0.645 | 0.005 | 0.645 | 0.005 |
11 | SE–SAA | - | - | 0.713 | 0.006 | 0.713 | 0.006 |
12 | SE–SS | - | - | 0.514 | 0.006 | 0.514 | 0.006 |
13 | SE–SPE | - | - | 0.615 | 0.006 | 0.615 | 0.006 |
14 | SI–SAU | 0.873 | 0.007 | - | - | 0.873 | 0.007 |
15 | SI–SP | 0.765 | 0.005 | - | - | 0.765 | 0.005 |
16 | SI–SAA | - | - | 0.845 | 0.004 | 0.845 | 0.004 |
17 | SI–SS | - | - | 0.610 | 0.004 | 0.610 | 0.004 |
18 | SI–SPE | - | - | 0.730 | 0.007 | 0.730 | 0.007 |
19 | SAU–SP | - | - | - | - | - | - |
20 | SAU–SAA | 0.307 | 0.052 | - | - | 0.307 | 0.052 |
21 | SAU–SS | 0.699 | 0.008 | - | - | 0.699 | 0.008 |
22 | SAU–SPE | - | - | 0.680 | 0.011 | 0.680 | 0.011 |
23 | SP–SAA | 0.754 | 0.014 | - | - | 0.754 | 0.014 |
24 | SP–SS | - | - | - | - | - | - |
25 | SP–SPE | - | - | 0.179 | 0.018 | 0.179 | 0.018 |
26 | SAA–SS | - | - | - | - | - | - |
27 | SAA–SPE | 0.237 | 0.024 | - | - | 0.237 | 0.024 |
28 | SS–SPE | 0.868 | 0.009 | - | - | 0.868 | 0.009 |
Table 6 shows that the five parameters, namely the Incremental Fit Index, Tucker Lewis Index, the Comparative Fit Index, Goodness of Fit Index, and Adjusted Goodness Fit Index, are all acceptable with parameter estimates greater than 0.8, whereas mean square error is excellent with parameter estimates less than 0.08.
4. Conclusions
The education system has been affected by the 2019 coronavirus disease; face-to-face classes are suspended to control and reduce the spread of the virus and infections [ 2 ]. The suspension of face-to-face classes results in the application of modular distance learning for K-12 students according to continuity of learning during the COVID-19 pandemic. With the outbreak of COVID-19, some issues concerning students’ academic Performance and factors associated with students’ psychological status are starting to emerge, which impacted the students’ ability to learn. This study aimed to perceive the impact of Modular Distance learning on the K-12 students amid the COVID-19 pandemic and assess cognitive factors affecting students’ academic achievement and satisfaction.
This study applied Transactional Distance Theory (TDT) and Bloom Taxonomy Theory (BTT) to evaluate cognitive factors affecting students’ academic achievements and satisfaction and evaluate the perceived effectiveness of K-12 students in response to modular learning. This study applied Structural Equation Modeling (SEM) to test hypotheses. The application of SEM analyzed the correlation among students’ background, experience, behavior, instructor interaction, performance, understanding, satisfaction, academic achievement, and student perceived effectiveness.
A total of 252 data samples were gathered through an online questionnaire. Based on findings, this study concludes that students’ background in modular distance learning affects their behavior and experience. Students’ experiences had significant effects on the performance and understanding of students in modular distance learning. Student instructor interaction had a substantial impact on performance and learning; it explains how vital interaction with the instructor is. The student interacting with the instructor shows that the student may receive feedback and guidance from the instructor. Understanding has a significant influence on students’ satisfaction and academic achievement. Student performance has a substantial impact on students’ academic achievement and satisfaction. Perceived effectiveness was significantly influenced by students’ academic achievement and student satisfaction. However, students’ behavior had no considerable effect on students’ instructor interaction, and students’ understanding while student performance equally had no significant impact on student satisfaction. From this study, students are likely to manifest good performance, behavior, and cognition when they have prior knowledge with regard to modular distance learning. This study will help the government, teachers, and students take the necessary steps to improve and enhance modular distance learning that will benefit students for effective learning.
The modular learning system has been in place since its inception. One of its founding metaphoric pillars is student satisfaction with modular learning. The organization demonstrated its dedication to the student’s voice as a component of understanding effective teaching and learning. Student satisfaction research has been transformed by modular learning. It has caused the education research community to rethink long-held assumptions that learning occurs primarily within a metaphorical container known as a “course.” When reviewing studies on student satisfaction from a factor analytic perspective, one thing becomes clear: this is a complex system with little consensus. Even the most recent factor analytical studies have done little to address the lack of understanding of the dimensions underlying satisfaction with modular learning. Items about student satisfaction with modular distance learning correspond to forming a psychological contract in factor analytic studies. The survey responses are reconfigured into a smaller number of latent (non-observable) dimensions that the students never really articulate but are fully expected to satisfy. Of course, instructors have contracts with their students. Studies such as this one identify the student’s psychological contact after the fact, rather than before the class. The most important aspect is the rapid adoption of this teaching and learning mode in Senior High School. Another balancing factor is the growing sense of student agency in the educational process. Students can express their opinions about their educational experiences in formats ranging from end-of-course evaluation protocols to various social networks, making their voices more critical.
Furthermore, they all agreed with latent trait theory, which holds that the critical dimensions that students differentiate when expressing their opinions about modular learning are formed by the combination of the original items that cannot be directly observed—which underpins student satisfaction. As stated in the literature, the relationship between student satisfaction and the characteristic of a psychological contract is illustrated. Each element is translated into how it might be expressed in the student’s voice, and then a contract feature and an assessment strategy are added. The most significant contributor to the factor pattern, engaged learning, indicates that students expect instructors to play a facilitative role in their teaching. This dimension corresponds to the relational contract, in which the learning environment is stable and well organized, with a clear path to success.
5. Limitations and Future Work
This study was focused on the cognitive capacity of modular distance learning towards academic achievements and satisfaction of K-12 students during the COVID-19 pandemic. The sample size in this study was small, at only 252. If this study is repeated with a larger sample size, it will improve the results. The study’s restriction was to the province of Occidental Mindoro; Structural Equation Modeling (SEM) was used to measure all the variables. Thus, this will give an adequate solution to the problem in the study.
The current study underlines that combining TDT and BTT can positively impact the research outcome. The contribution the current study might make to the field of modular distance learning has been discussed and explained. Based on this research model, the nine (9) factors could broadly clarify the students’ adoption of new learning environment platform features. Thus, the current research suggests that more investigation be carried out to examine relationships among the complexity of modular distance learning.
Funding Statement
This research received no external funding.
Author Contributions
Data collection, methodology, writing and editing, K.A.M.; data collection, writing—review and editing, Y.-T.J. and C.S.S. All authors have read and agreed to the published version of the manuscript.
Institutional Review Board Statement
Informed consent statement.
Informed consent was obtained from all subjects involved in the study.
Data Availability Statement
Conflicts of interest.
The authors declare no conflict of interest.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
![research paper about modular distance learning research paper about modular distance learning](https://himjournals.com/media/journal_thumbnails/HJEL_y5i3tCe.png)
- All Archives
- Special Issue
- Aim and Scope
- Editorial Board
- Journal Insights
- Abstracting and Indexing
- Announcements
- For Authors
- For Reviewers
- Ethics and Policies
- Open Access Policies
- Plagiarism Policies
- Research Ethics Policy
- Authorship and Contributorship
- Informed Consent Policy
- Conflict of Interest
- Content Licensing
- Corrections and Retractions
- Complaints and Appeals
- Article Submission
![research paper about modular distance learning research paper about modular distance learning](https://himjournals.com/static/images/icons/quote.png)
Download PDF
![research paper about modular distance learning Chat on WhatsApp](https://himjournals.com/static/images/icons/whatsapp.png)
Perceived Effects of Modular Distance Learning on the Students in Aumbay Elementary School
International Journal of Research and Innovation in Social Science (IJRISS) 2022, 6(9), 496-506
11 Pages Posted: 20 Jan 2023
Livelyn Argallon
Um peñaplata college, jaycee rylle loreño, honey jane paham, jovenil bacatan.
UM Peñaplata College; The Rizal Memorial Colleges; Holy Cross of Davao College
Date Written: September 2022
This research aimed to gather data on the perceived effects of modular distance learning on students. The following research objectives informed this research: 1) to determine the perceived effects of modular distance learning on the students and 2) to determine the significant difference in the perceived effects of modular distance learning on grades 4,5 and 6 students. This quantitative study employed a descriptive survey to determine and identify the perceived effects of modular distance learning on grades 4,5 and 6 students. The research instrument used in this study is a survey questionnaire, which is used to acquire the necessary data. Based on the result, the study revealed that students agreed on using modular distance learning as a teaching method. They agreed most with the statements and the benefits of this approach. The study also revealed no significant difference in the perceived effects of modular distance learning on students when categorized according to gender. This means that students had the same perceptions when categorized according to gender. However, the study also revealed a significant difference in the perceived effects of modular distance learning on the students when categorized according to their grade level. It means that students had a different perception when they were grouped according to their grade level. Therefore, researchers concluded that in terms of gender, modular distance learning was not affected by their perception of the modular approach. At the same time, the perception of the students had a difference in terms of their grade level. This study implies that this modular distance learning approach helps students improve their perception of modules. Therefore, a modular approach must be continued in face-to-face instruction even if the pandemic vanishes. Researchers may recommend to the faculty/teachers in Aumbay Elementary School to use modules so students can study in advance; teachers/faculty might monitor their student’s progress and be more flexible and even-minded in any situation to provide and help the students. Moreover, Schools may provide guidelines regarding the implementation of limited face-to-face and modular distance learning so that students can learn in different approaches; parents should also give attention to their children and support them in such activities. Furthermore, future researchers may incorporate other variables when they conduct a study related to our research.
Keywords: modular approach, perceived effects, modular distance learning, students
JEL Classification: Y80
Suggested Citation: Suggested Citation
Jovenil Bacatan (Contact Author)
Um peñaplata college ( email ).
Penaplata Island Garden City of Samal, Davao del Norte 8119 Philippines
The Rizal Memorial Colleges ( email )
Davao City, Davao del Sur 8000 Philippines
Holy Cross of Davao College ( email )
Sta. Ana Avenue Davao City, Davao del Sur 8000 Philippines
Do you have a job opening that you would like to promote on SSRN?
Paper statistics, related ejournals, pedagogy ejournal.
Subscribe to this fee journal for more curated articles on this topic
Cognitive Psychology eJournal
Developmental psychology ejournal.
- DOI: 10.12973/eu-jer.11.2.917
- Corpus ID: 247015616
Impact of Modular Distance Learning on High School Students Mathematics Motivation, Interest/Attitude, Anxiety and Achievement during the COVID-19 Pandemic
- Andie Tangonan
- Published in European Journal of… 15 April 2022
- Mathematics, Education
Tables from this paper
![research paper about modular distance learning table 1](https://d3i71xaburhd42.cloudfront.net/c3c88b5919ab6c1d98188b1d26dbeb415524f3c1/4-Table1-1.png)
16 Citations
The effect of teaching conceptual knowledge on students’ achievement, anxiety about, and attitude toward mathematics.
- Highly Influenced
It Doesn't Mean that Students Don't Have Mathematics Anxiety: A Case Study of Mathematics Learning with Path Analysis
Undergraduate students’ modular experiences in learning mathematics in the new normal education, enhancing readiness and mathematics performance in face-to-face learning among grade 7 students: a descriptive-correlational study, analysis of learning interest, level of discipline, and student’s responsibility for post covid-19 learning outcomes, international journal of evaluation and research in education (ijere), using gamification to motivate students with simple-moderate intellectual disabilities, european journal of educational research, exploring factors influencing student's learning difficulties during pandemic in indonesia: a structural equation modelling, multisensory supplementary instructional material in earth and space science, 71 references, students’ mathematics self-concept, mathematics anxiety and mathematics self-regulated learning during the covid-19 pandemic.
- Highly Influential
Mathematics Anxiety and Achievement Among Secondary School Students
Association of interest, attitude and learning habit in mathematics learning towards enhancing students’ achievement, effects of gender, mathematics anxiety and achievement motivation on college students’ achievement in mathematics, attitudes, study habits, and academic performance of junior high school students in mathematics, student learning, childhood & voices | research article creativity in unique problem-solving in mathematics and its influence on motivation for learning, students’ motivation in the mathematics classroom. revealing causes and consequences, comparison on the effectiveness of modular learning in general mathematics among the senior high school strands, relationship between interest and mathematics performance in a technology-enhanced learning context in malaysia, relations between students’ mathematics anxiety and motivation to learn mathematics: a meta-analysis, related papers.
Showing 1 through 3 of 0 Related Papers
- Teaching Methods
- Distance Learning
Perceptions, Challenges and Effectiveness of Modular Distance Learning Approach to The Academic Performance of Humanities and Social Sciences (HUMSS) Students of Botolan National High School
- International Journal of Multidisciplinary Applied Business and Education Research 4(6):1833-1848
- 4(6):1833-1848
![research paper about modular distance learning Norvin De Ocampo at Department of Education of the Philippines](https://c5.rgstatic.net/m/4671872220764/images/template/default/profile/profile_default_m.jpg)
- Department of Education of the Philippines
Abstract and Figures
![research paper about modular distance learning Distribution of Student-Respondents per Section](https://www.researchgate.net/publication/371812611/figure/tbl1/AS:11431281170067777@1687608602208/Distribution-of-Student-Respondents-per-Section_Q320.jpg)
Discover the world's research
- 25+ million members
- 160+ million publication pages
- 2.3+ billion citations
- Agnes Bitong
![research paper about modular distance learning Maria Arvelyn Beran Rios](https://i1.rgstatic.net/ii/profile.image/11431281214100988-1703382221623_Q64/Maria-Arvelyn-Rios.jpg)
- Narasimha Paravastu
- Sam S. Ramanujan
![research paper about modular distance learning Dafid Slamet Setiana](https://c5.rgstatic.net/m/4671872220764/images/template/default/profile/profile_default_m.jpg)
- T O Adeyemi
- A Kamaludin
- M A Almaiah
- A Al-Khasawneh
- A Althunibat
- E Alqurashi
- Recruit researchers
- Join for free
- Login Email Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google Welcome back! Please log in. Email · Hint Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google No account? Sign up
Academia.edu no longer supports Internet Explorer.
To browse Academia.edu and the wider internet faster and more securely, please take a few seconds to upgrade your browser .
- We're Hiring!
- Help Center
Modular Distance Learning
- Most Cited Papers
- Most Downloaded Papers
- Newest Papers
Enter the email address you signed up with and we'll email you a reset link.
- Academia.edu Journals
- We're Hiring!
- Help Center
- Find new research papers in:
- Health Sciences
- Earth Sciences
- Cognitive Science
- Mathematics
- Computer Science
- Academia ©2024
Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.
- View all journals
- Explore content
- About the journal
- Publish with us
- Sign up for alerts
- Open access
- Published: 21 August 2024
Loss of plasticity in deep continual learning
- Shibhansh Dohare ORCID: orcid.org/0000-0002-3796-9347 1 ,
- J. Fernando Hernandez-Garcia 1 ,
- Qingfeng Lan ORCID: orcid.org/0000-0001-8568-7603 1 ,
- Parash Rahman ORCID: orcid.org/0000-0001-8430-2679 1 ,
- A. Rupam Mahmood ORCID: orcid.org/0000-0001-6266-162X 1 , 2 &
- Richard S. Sutton 1 , 2
Nature volume 632 , pages 768–774 ( 2024 ) Cite this article
29k Accesses
181 Altmetric
Metrics details
- Computer science
- Human behaviour
Artificial neural networks, deep-learning methods and the backpropagation algorithm 1 form the foundation of modern machine learning and artificial intelligence. These methods are almost always used in two phases, one in which the weights of the network are updated and one in which the weights are held constant while the network is used or evaluated. This contrasts with natural learning and many applications, which require continual learning. It has been unclear whether or not deep learning methods work in continual learning settings. Here we show that they do not—that standard deep-learning methods gradually lose plasticity in continual-learning settings until they learn no better than a shallow network. We show such loss of plasticity using the classic ImageNet dataset and reinforcement-learning problems across a wide range of variations in the network and the learning algorithm. Plasticity is maintained indefinitely only by algorithms that continually inject diversity into the network, such as our continual backpropagation algorithm, a variation of backpropagation in which a small fraction of less-used units are continually and randomly reinitialized. Our results indicate that methods based on gradient descent are not enough—that sustained deep learning requires a random, non-gradient component to maintain variability and plasticity.
Similar content being viewed by others
![research paper about modular distance learning research paper about modular distance learning](https://media.springernature.com/w215h120/springer-static/image/art%3A10.1038%2Fs41593-023-01514-1/MediaObjects/41593_2023_1514_Fig1_HTML.png)
Inferring neural activity before plasticity as a foundation for learning beyond backpropagation
![research paper about modular distance learning research paper about modular distance learning](https://media.springernature.com/w215h120/springer-static/image/art%3A10.1038%2Fs42256-022-00568-3/MediaObjects/42256_2022_568_Fig1_HTML.png)
Three types of incremental learning
![research paper about modular distance learning research paper about modular distance learning](https://media.springernature.com/w215h120/springer-static/image/art%3A10.1038%2Fs41467-023-37562-1/MediaObjects/41467_2023_37562_Fig1_HTML.png)
Meta-learning biologically plausible plasticity rules with random feedback pathways
Machine learning and artificial intelligence have made remarkable progress in the past decade, with landmark successes in natural-language processing 2 , 3 , biology 4 , game playing 5 , 6 , 7 , 8 and robotics 9 , 10 . All these systems use artificial neural networks, whose computations are inspired by the operation of human and animal brains. Learning in these networks refers to computational algorithms for changing the strengths of their connection weights (computational synapses). The most important modern learning methods are based on stochastic gradient descent (SGD) and the backpropagation algorithm, ideas that originated at least four decades ago but are much more powerful today because of the availability of vastly greater computer power. The successes are also because of refinements of the learning and training techniques that together make the early ideas effective in much larger and more deeply layered networks. These methodologies are collectively referred to as deep learning.
Despite its successes, deep learning has difficulty adapting to changing data. Because of this, in almost all applications, deep learning is restricted to a special training phase and then turned off when the network is actually used. For example, large language models such as ChatGPT are trained on a large generic training set and then fine-tuned on smaller datasets specific to an application or to meet policy and safety goals, but finally their weights are frozen before the network is released for use. With current methods, it is usually not effective to simply continue training on new data when they become available. The effect of the new data is either too large or too small and not properly balanced with the old data. The reasons for this are not well understood and there is not yet a clear solution. In practice, the most common strategy for incorporating substantial new data has been simply to discard the old network and train a new one from scratch on the old and new data together 11 , 12 . When the network is a large language model and the data are a substantial portion of the internet, then each retraining may cost millions of dollars in computation. Moreover, a wide range of real-world applications require adapting to change. Change is ubiquitous in learning to anticipate markets and human preferences and in gaming, logistics and control systems. Deep-learning systems would be much more powerful if they, like natural-learning systems, were capable of continual learning.
Here we show systematically that standard deep-learning methods lose their ability to learn with extended training on new data, a phenomenon that we call loss of plasticity. We use classic datasets, such as ImageNet and CIFAR-100, modified for continual learning, and standard feed-forward and residual networks with a wide variety of standard learning algorithms. Loss of plasticity in artificial neural networks was first shown at the turn of the century in the psychology literature 13 , 14 , 15 , before the development of deep-learning methods. Plasticity loss with modern methods was visible in some recent works 11 , 16 , 17 , 18 and most recently has begun to be explored explicitly 12 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 . Loss of plasticity is different from catastrophic forgetting, which concerns poor performance on old examples even if they are not presented again 28 , 29 , 30 .
Although standard deep-learning methods lose plasticity with extended learning, we show that a simple change enables them to maintain plasticity indefinitely in both supervised and reinforcement learning. Our new algorithm, continual backpropagation, is exactly like classical backpropagation except that a tiny proportion of less-used units are reinitialized on each step much as they were all initialized at the start of training. Continual backpropagation is inspired by a long history of methods for automatically generating and testing features, starting with Selfridge’s Pandemonium in 1959 (refs. 19 , 20 , 31 , 32 , 33 , 34 , 35 ). The effectiveness of continual backpropagation shows that the problem of plasticity loss is not inherent in artificial neural networks.
Plasticity loss in supervised learning
The primary purpose of this article is to demonstrate loss of plasticity in standard deep-learning systems. For the demonstration to be convincing, it must be systematic and extensive. It must consider a wide range of standard deep-learning networks, learning algorithms and parameter settings. For each of these, the experiments must be run long enough to expose long-term plasticity loss and be repeated enough times to obtain statistically significant results. Altogether, more computation is needed by three or four orders of magnitude compared with what would be needed to train a single network. For example, a systematic study with large language models would not be possible today because just a single training run with one of these networks would require computation costing millions of dollars. Fortunately, advances in computer hardware have continued apace since the development of deep learning and systematic studies have become possible with the deep-learning networks used earlier and with some of the longer-lived test problems. Here we use ImageNet, a classic object-recognition test bed 36 , which played a pivotal role in the rise of deep learning 37 and is still influential today.
The ImageNet database comprises millions of images labelled by nouns (classes) such as types of animal and everyday object. The typical ImageNet task is to guess the label given an image. The standard way to use this dataset is to partition it into training and test sets. A learning system is first trained on a set of images and their labels, then training is stopped and performance is measured on a separate set of test images from the same classes. To adapt ImageNet to continual learning while minimizing all other changes, we constructed a sequence of binary classification tasks by taking the classes in pairs. For example, the first task might be to distinguish cats from houses and the second might be to distinguish stop signs from school buses. With the 1,000 classes in our dataset, we were able to form half a million binary classification tasks in this way. For each task, a deep-learning network was first trained on a subset of the images for the two classes and then its performance was measured on a separate test set for the classes. After training and testing on one task, the next task began with a different pair of classes. We call this problem ‘Continual ImageNet’. In Continual ImageNet, the difficulty of tasks remains the same over time. A drop in performance would mean the network is losing its learning ability, a direct demonstration of loss of plasticity.
We applied a wide variety of standard deep-learning networks to Continual ImageNet and tested many learning algorithms and parameter settings. To assess the performance of the network on a task, we measured the percentage of test images that were correctly classified. The results shown in Fig. 1b are representative; they are for a feed-forward convolutional network and for a training procedure, using unmodified backpropagation, that performed well on this problem in the first few tasks.
![research paper about modular distance learning figure 1](https://media.springernature.com/lw685/springer-static/image/art%3A10.1038%2Fs41586-024-07711-7/MediaObjects/41586_2024_7711_Fig1_HTML.png)
a – c , In a sequence of binary classification tasks using ImageNet pictures ( a ), the conventional backpropagation algorithm loses plasticity at all step sizes ( b ), whereas the continual backpropagation, L2 regularization and Shrink and Perturb algorithms maintain plasticity, apparently indefinitely ( c ). All results are averaged over 30 runs; the solid lines represent the mean and the shaded regions correspond to ±1 standard error.
Although these networks learned up to 88% correct on the test set of the early tasks (Fig. 1b , left panel), by the 2,000th task, they had lost substantial plasticity for all values of the step-size parameter (right panel). Some step sizes performed well on the first two tasks but then much worse on subsequent tasks, eventually reaching a performance level below that of a linear network. For other step sizes, performance rose initially and then fell and was only slightly better than the linear network after 2,000 tasks. We found this to be a common pattern in our experiments: for a well-tuned network, performance first improves and then falls substantially, ending near or below the linear baseline. We have observed this pattern for many network architectures, parameter choices and optimizers. The specific choice of network architecture, algorithm parameters and optimizers affected when the performance started to drop, but a severe performance drop occurred for a wide range of choices. The failure of standard deep-learning methods to learn better than a linear network in later tasks is direct evidence that these methods do not work well in continual-learning problems.
Algorithms that explicitly keep the weights of the network small were an exception to the pattern of failure and were often able to maintain plasticity and even improve their performance over many tasks, as shown in Fig. 1c . L2 regularization adds a penalty for large weights; augmenting backpropagation with this enabled the network to continue improving its learning performance over at least 5,000 tasks. The Shrink and Perturb algorithm 11 , which includes L2 regularization, also performed well. Best of all was our continual backpropagation algorithm, which we discuss later. For all algorithms, we tested a wide range of parameter settings and performed many independent runs for statistical significance. The presented curves are the best representative of each algorithm.
For a second demonstration, we chose to use residual networks, class-incremental continual learning and the CIFAR-100 dataset. Residual networks include layer-skipping connections as well as the usual layer-to-layer connections of conventional convolutional networks. The residual networks of today are more widely used and produce better results than strictly layered networks 38 . Class-incremental continual learning 39 involves sequentially adding new classes while testing on all classes seen so far. In our demonstration, we started with training on five classes and then successively added more, five at a time, until all 100 were available. After each addition, the networks were trained and performance was measured on all available classes. We continued training on the old classes (unlike in most work in class-incremental learning) to focus on plasticity rather than on forgetting.
In this demonstration, we used an 18-layer residual network with a variable number of heads, adding heads as new classes were added. We also used further deep-learning techniques, including batch normalization, data augmentation, L2 regularization and learning-rate scheduling. These techniques are standardly used with residual networks and are necessary for good performance. We call this our base deep-learning system.
As more classes are added, correctly classifying images becomes more difficult and classification accuracy would decrease even if the network maintained its ability to learn. To factor out this effect, we compare the accuracy of our incrementally trained networks with networks that were retrained from scratch on the same subset of classes. For example, the network that was trained first on five classes, and then on all ten classes, is compared with a network retrained from scratch on all ten classes. If the incrementally trained network performs better than a network retrained from scratch, then there is a benefit owing to training on previous classes, and if it performs worse, then there is genuine loss of plasticity.
The red line in Fig. 2b shows that incremental training was initially better than retraining, but after 40 classes, the incrementally trained network showed loss of plasticity that became increasingly severe. By the end, when all 100 classes were available, the accuracy of the incrementally trained base system was 5% lower than the retrained network (a performance drop equivalent to that of removing a notable algorithmic advance, such as batch normalization). Loss of plasticity was less severe when Shrink and Perturb was added to the learning algorithm (in the incrementally trained network) and was eliminated altogether when continual backpropagation (see the ‘Maintaining plasticity through variability and selective preservation’ section) was added. These additions also prevented units of the network from becoming inactive or redundant, as shown in Fig. 2c,d .
![research paper about modular distance learning figure 2](https://media.springernature.com/lw685/springer-static/image/art%3A10.1038%2Fs41586-024-07711-7/MediaObjects/41586_2024_7711_Fig2_HTML.png)
a , An incrementally growing image-classification problem. b , Initially, accuracy is improved by incremental training compared with a network trained from scratch, but after 40 classes, accuracy degrades substantially in a base deep-learning system, less so for a Shrink and Perturb learning system and not at all for a learning system based on continual backpropagation. c , The number of network units that are active less than 1% of the time increases rapidly for the base deep-learning system, but less so for Shrink and Perturb and continual backpropagation systems. d , A low stable rank means that the units of a network do not provide much diversity; the base deep-learning system loses much more diversity than the Shrink and Perturb and continual backpropagation systems. All results are averaged over 30 runs; the solid lines represent the mean and the shaded regions correspond to ±1 standard error.
This demonstration involved larger networks and required more computation, but still we were able to perform extensive systematic tests. We found a robust pattern in the results that was similar to what we found in ImageNet. In both cases, deep-learning networks exhibited substantial loss of plasticity. Altogether, these results, along with other extensive results in Methods , constitute substantial evidence of plasticity loss.
Plasticity loss in reinforcement learning
Continual learning is essential to reinforcement learning in ways that go beyond its importance in supervised learning. Not only can the environment change but the behaviour of the learning agent can also change, thereby influencing the data it receives even if the environment is stationary. For this reason, the need for continual learning is often more apparent in reinforcement learning, and reinforcement learning is an important setting in which to demonstrate the tendency of deep learning towards loss of plasticity.
Nevertheless, it is challenging to demonstrate plasticity loss in reinforcement learning in a systematic and rigorous way. In part, this is because of the great variety of algorithms and experimental settings that are commonly used in reinforcement-learning research. Algorithms may learn value functions, behaviours or both simultaneously and may involve replay buffers, world models and learned latent states. Experiments may be episodic, continuing or offline. All of these choices involve several embedded choices of parameters. More fundamentally, reinforcement-learning algorithms affect the data seen by the agent. The learning ability of an algorithm is thus confounded with its ability to generate informative data. Finally, and in part because of the preceding, reinforcement-learning results tend to be more stochastic and more widely varying than in supervised learning. Altogether, demonstration of reinforcement-learning abilities, particularly negative results, tends to require more runs and generally much more experimental work and thus inevitably cannot be as definitive as in supervised learning.
Our first demonstration involves a reinforcement-learning algorithm applied to a simulated ant-like robot tasked with moving forwards as rapidly and efficiently as possible. The agent–environment interaction comprises a series of episodes, each beginning in a standard state and lasting up to 1,000 time steps. On each time step, the agent receives a reward depending on the forward distance travelled and the magnitude of its action (see Methods for details). An episode terminates in fewer than 1,000 steps if the ant jumps too high instead of moving forwards, as often happens early in learning. In the results to follow, we use the cumulative reward during an episode as our primary performance measure. To make the task non-stationary (and thereby emphasize plasticity), the coefficient of friction between the feet of the ant and the floor is changed after every 2 million time steps (but only at an episode boundary; details in Methods ). For fastest walking, the agent must adapt (relearn) its way of walking each time the friction changes. For this experiment, we used the proximal policy optimization (PPO) algorithm 40 . PPO is a standard deep reinforcement-learning algorithm based on backpropagation. It is widely used, for example, in robotics 9 , in playing real-time strategy games 41 and in aligning large language models from human feedback 42 .
PPO performed well (see the red line in Fig. 3c ) for the first 2 million steps, up until the first change in friction, but then performed worse and worse. Note how the performance of the other algorithms in Fig. 3c decreased each time the friction changed and then recovered as the agent adapted to the new friction, giving the plot a sawtooth appearance. PPO augmented with a specially tuned Adam optimizer 24 , 43 performed much better (orange line in Fig. 3c ) but still performed much worse over successive changes after the first two, indicating substantial loss of plasticity. On the other hand, PPO augmented with L2 regularization and continual backpropagation largely maintained their plasticity as the problem changed.
![research paper about modular distance learning figure 3](https://media.springernature.com/lw685/springer-static/image/art%3A10.1038%2Fs41586-024-07711-7/MediaObjects/41586_2024_7711_Fig3_HTML.png)
a , The reinforcement-learning agent controls torques at the eight joints of the simulated ant (red circles) to maximize forward motion and minimize penalties. b , Here we use a version of the ant problem in which the friction on contact with the ground is abruptly changed every 2 million time steps. c , The standard PPO learning algorithm fails catastrophically on the non-stationary ant problem. If the optimizer of PPO (Adam) is tuned in a custom way, then the failure is less severe, but adding continual backpropagation or L2 regularization is necessary to perform well indefinitely. These results are averaged over 100 runs; the solid lines represent the mean and the shaded regions represent the 95% bootstrapped confidence interval.
Now consider the same ant-locomotion task except with the coefficient of friction held constant at an intermediate value over 50 million time steps. The red line in Fig. 4a shows that the average performance of PPO increased for about 3 million steps but then collapsed. After 20 million steps, the ant is failing every episode and is unable to learn to move forwards efficiently. The red lines in the other panels of Fig. 4 provide further insight into the loss of plasticity of PPO. They suggest that the network may be losing plasticity in the same way as in our supervised learning results (see Fig. 2 and Extended Data Fig. 3c ). In both cases, most of the network’s units became dormant during the experiment, and the network markedly lost stable rank. The addition of L2 regularization mitigated the performance degradation by preventing continual growth of weights but also resulted in very small weights (Fig. 4d ), which prevented the agent from committing to good behaviour. The addition of continual backpropagation performed better overall. We present results for continual backpropagation only with (slight) L2 regularization, because without it, performance was highly sensitive to parameter settings. These results show that plasticity loss can be catastrophic in both deep reinforcement learning as well as deep supervised learning.
![research paper about modular distance learning figure 4](https://media.springernature.com/lw685/springer-static/image/art%3A10.1038%2Fs41586-024-07711-7/MediaObjects/41586_2024_7711_Fig4_HTML.png)
a , The four reinforcement-learning algorithms performed similarly on this and the non-stationary problem (compare with Fig. 3c ). b , c , A closer look inside the networks reveals a similar pattern as in supervised learning (compare with Fig. 2c,d ). d , The absolute values of the weights of the networks increased steadily under standard and tuned PPO, whereas they decreased and stayed small under L2 regularization with or without continual backpropagation. These results are averaged over 30 runs; the solid lines represent the mean and the shaded regions represent the 95% bootstrapped confidence interval.
Maintaining plasticity
Surprisingly, popular methods such as Adam, Dropout and normalization actually increased loss of plasticity (see Extended Data Fig. 4a ). L2 regularization, on the other hand, reduced loss of plasticity in many cases (purple line in Figs. 1 , 3 and 4 ). L2 regularization stops the weights from becoming too large by moving them towards zero at each step. The small weights allow the network to remain plastic. Another existing method that reduced loss of plasticity is Shrink and Perturb 11 (orange line in Figs. 1 and 2 ). Shrink and Perturb is L2 regularization plus small random changes in weights at each step. The injection of variability into the network can reduce dormancy and increase the diversity of the representation (Figs. 2 and 4 ). Our results indicate that non-growing weights and sustained variability in the network may be important for maintaining plasticity.
We now describe a variation of the backpropagation algorithm that is explicitly designed to inject variability into the network and keep some of its weights small. Conventional backpropagation has two main parts: initialization with small random weights before training and then gradient descent at each training step. The initialization provides variability initially, but, as we have seen, with continued training, variability tends to be lost, as well as plasticity along with it. To maintain the variability, our new algorithm, continual backpropagation, reinitializes a small number of units during training, typically fewer than one per step. To prevent disruption of what the network has already learned, only the least-used units are considered for reinitialization. See Methods for details.
The blue line in Fig. 1c shows the performance of continual backpropagation on Continual ImageNet. It mitigated loss of plasticity in Continual ImageNet while outperforming existing methods. Similarly, the blue lines in Fig. 2 show the performance of continual backpropagation on class-incremental CIFAR-100 and its effect on the evolution of dormant units and stable rank. Continual backpropagation fully overcame loss of plasticity, with a high stable rank and almost no dead units throughout learning.
In reinforcement learning, continual backpropagation was applied together with L2 regularization (a small amount of regularization was added to prevent excessive sensitivity to parameters in reinforcement-learning experiments). The blue line in Fig. 3 shows the performance of PPO with continual backpropagation on the ant-locomotion problem with changing friction. PPO with continual backpropagation performed much better than standard PPO, with little or no loss of plasticity. On the ant-locomotion problem with constant friction (Fig. 4 ), PPO with continual backpropagation continued improving throughout the experiment. The blue lines in Fig. 4b–d show the evolution of the correlates of loss of plasticity when we used continual backpropagation. PPO with continual backpropagation had few dormant units, a high stable rank and an almost constant average weight magnitude.
Our results are consistent with the idea that small weights reduce loss of plasticity and that a continual injection of variability further mitigates loss of plasticity. Although Shrink and Perturb adds variability to all weights, continual backpropagation does so selectively, and this seems to enable it to better maintain plasticity. Continual backpropagation involves a form of variation and selection in the space of neuron-like units, combined with continuing gradient descent. The variation and selection is reminiscent of trial-and-error processes in evolution and behaviour 44 , 45 , 46 , 47 and has precursors in many earlier ideas, including Keifer–Wolfowitz methods 48 and restart methods 49 in engineering and feature-search methods in machine learning 31 , 32 , 33 , 34 , 35 , 50 . Continual backpropagation brings a form of this old idea to modern deep learning. However, it is just one variation of this idea; other variations are possible and some of these have been explored in recent work 25 , 27 . We look forward to future work that explicitly compares and further refines these variations.
Deep learning is an effective and valuable technology in settings in which learning occurs in a special training phase and not thereafter. In settings in which learning must continue, however, we have shown that deep learning does not work. By deep learning, we mean the existing standard algorithms for learning in multilayer artificial neural networks and by not work, we mean that, over time, they fail to learn appreciably better than shallow networks. We have shown such loss of plasticity using supervised-learning datasets and reinforcement-learning tasks on which deep learning has previously excelled and for a wide range of networks and standard learning algorithms. Taking a closer look, we found that, during training, many of the networks’ neuron-like units become dormant, overcommitted and similar to each other, hampering the ability of the networks to learn new things. As they learn, standard deep-learning networks gradually and irreversibly lose their diversity and thus their ability to continue learning. Plasticity loss is often severe when learning continues for many tasks, but may not occur at all for small numbers of tasks.
The problem of plasticity loss is not intrinsic to deep learning. Deep artificial neural networks trained by gradient descent are perfectly capable of maintaining their plasticity, apparently indefinitely, as we have shown with the Shrink and Perturb algorithm and particularly with the new continual backpropagation algorithm. Both of these algorithms extend standard deep learning by adding a source of continuing variability to the weights of the network, and continual backpropagation restricts this variability to the units of the network that are at present least used, minimizing damage to the operation of the network. That is, continual backpropagation involves a form of variation and selection in the space of neuron-like units, combined with continuing gradient descent. This idea has many historical antecedents and will probably require further development to reach its most effective form.
Specifics of continual backpropagation
Continual backpropagation selectively reinitializes low-utility units in the network. Our utility measure, called the contribution utility, is defined for each connection or weight and each unit. The basic intuition behind the contribution utility is that the magnitude of the product of units’ activation and outgoing weight gives information about how valuable this connection is to its consumers. If the contribution of a hidden unit to its consumer is small, its contribution can be overwhelmed by contributions from other hidden units. In such a case, the hidden unit is not useful to its consumer. We define the contribution utility of a hidden unit as the sum of the utilities of all its outgoing connections. The contribution utility is measured as a running average of instantaneous contributions with a decay rate, η , which is set to 0.99 in all experiments. In a feed-forward neural network, the contribution utility, u l [ i ], of the i th hidden unit in layer l at time t is updated as
in which h l , i , t is the output of the i th hidden unit in layer l at time t , w l , i , k , t is the weight connecting the i th unit in layer l to the k th unit in layer l + 1 at time t and n l +1 is the number of units in layer l + 1.
When a hidden unit is reinitialized, its outgoing weights are initialized to zero. Initializing the outgoing weights as zero ensures that the newly added hidden units do not affect the already learned function. However, initializing the outgoing weight to zero makes the new unit vulnerable to immediate reinitialization, as it has zero utility. To protect new units from immediate reinitialization, they are protected from a reinitialization for maturity threshold m number of updates. We call a unit mature if its age is more than m . Every step, a fraction of mature units ρ , called the replacement rate, is reinitialized in every layer.
The replacement rate ρ is typically set to a very small value, meaning that only one unit is replaced after hundreds of updates. For example, in class-incremental CIFAR-100 (Fig. 2 ) we used continual backpropagation with a replacement rate of 10 −5 . The last layer of the network in that problem had 512 units. At each step, roughly 512 × 10 −5 = 0.00512 units are replaced. This corresponds to roughly one replacement after every 1/0.00512 ≈ 200 updates or one replacement after every eight epochs on the first five classes.
The final algorithm combines conventional backpropagation with selective reinitialization to continually inject random units from the initial distribution. Continual backpropagation performs a gradient descent and selective reinitialization step at each update. Algorithm 1 specifies continual backpropagation for a feed-forward neural network. In cases in which the learning system uses mini-batches, the instantaneous contribution utility can be used by averaging the utility over the mini-batch instead of keeping a running average to save computation (see Extended Data Fig. 5d for an example). Continual backpropagation overcomes the limitation of previous work 34 , 35 on selective reinitialization and makes it compatible with modern deep learning.
Algorithm 1
Continual backpropagation for a feed-forward network with L layers
Set replacement rate ρ , decay rate η and maturity threshold m
Initialize the weights w 0 ,…, w L −1 , in which w l is sampled from distribution d l
Initialize utilities u 1 ,…, u L −1 , number of units to replace c 1 ,…, c L −1 , and ages a 1 ,…, a L −1 to 0
For each input x t do
Forward pass: pass x t through the network to get the prediction \(\widehat{{{\bf{y}}}_{t}}\)
Evaluate: receive loss \(l({{\bf{x}}}_{t},\widehat{{{\bf{y}}}_{t}})\)
Backward pass: update the weights using SGD or one of its variants
For layer l in 1: L − 1 do
Update age: a l = a l + 1
Update unit utility: see equation ( 1 )
Find eligible units: n eligible = number of units with age greater than m
Update number of units to replace: c l = c l + n eligible × ρ
If c l > 1
Find the unit with smallest utility and record its index as r
Reinitialize input weights: resample w l −1 [:, r ] from distribution d l
Reinitialize output weights: set w l [ r ,:] to 0
Reinitialize utility and age: set u l [ r ] = 0 and a l [ r ] = 0
Update number of units to replace: c l = c l − 1
Details of Continual ImageNet
The ImageNet database we used consists of 1,000 classes, each of 700 images. The 700 images for each class were divided into 600 images for a training set and 100 images for a test set. On each binary classification task, the deep-learning network was first trained on the training set of 1,200 images and then its classification accuracy was measured on the test set of 200 images. The training consisted of several passes through the training set, called epochs. For each task, all learning algorithms performed 250 passes through the training set using mini-batches of size 100. All tasks used the downsampled 32 × 32 version of the ImageNet dataset, as is often done to save computation 51 .
All algorithms on Continual ImageNet used a convolutional network. The network had three convolutional-plus-max-pooling layers, followed by three fully connected layers, as detailed in Extended Data Table 3 . The final layer consisted of just two units, the heads, corresponding to the two classes. At task changes, the input weights of the heads were reset to zero. Resetting the heads in this way can be viewed as introducing new heads for the new tasks. This resetting of the output weights is not ideal for studying plasticity, as the learning system gets access to privileged information on the timing of task changes (and we do not use it in other experiments in this paper). We use it here because it is the standard practice in deep continual learning for this type of problem in which the learning system has to learn a sequence of independent tasks 52 .
In this problem, we reset the head of the network at the beginning of each task. It means that, for a linear network, the whole network is reset. That is why the performance of a linear network will not degrade in Continual ImageNet. As the linear network is a baseline, having a low-variance estimate of its performance is desirable. The value of this baseline is obtained by averaging over thousands of tasks. This averaging gives us a much better estimate of its performance than other networks.
The network was trained using SGD with momentum on the cross-entropy loss and initialized once before the first task. The momentum hyperparameter was 0.9. We tested various step-size parameters for backpropagation but only presented the performance for step sizes 0.01, 0.001 and 0.0001 for clarity of Fig. 1b . We performed 30 runs for each hyperparameter value, varying the sequence of tasks and other randomness. Across different hyperparameters and algorithms, the same sequences of pairs of classes were used.
We now describe the hyperparameter selection for L2 regularization, Shrink and Perturb and continual backpropagation. The main text presents the results for these algorithms on Continual ImageNet in Fig. 1c . We performed a grid search for all algorithms to find the set of hyperparameters that had the highest average classification accuracy over 5,000 tasks. The values of hyperparameters used for the grid search are described in Extended Data Table 2 . L2 regularization has two hyperparameters, step size and weight decay. Shrink and Perturb has three hyperparameters, step size, weight decay and noise variance. We swept over two hyperparameters of continual backpropagation: step size and replacement rate. The maturity threshold in continual backpropagation was set to 100. For both backpropagation and L2 regularization, the performance was poor for step sizes of 0.1 or 0.003. We chose to only use step sizes of 0.03 and 0.01 for continual backpropagation and Shrink and Perturb. We performed ten independent runs for all sets of hyperparameters. Then we performed another 20 runs to complete 30 runs for the best-performing set of hyperparameters to produce the results in Fig. 1c .
Class-incremental CIFAR-100
In the class-incremental CIFAR-100, the learning system gets access to more and more classes over time. Classes are provided to the learning system in increments of five. First, it has access to just five classes, then ten and so on, until it gets access to all 100 classes. The learning system is evaluated on the basis of how well it can discriminate between all the available classes at present. The dataset consists of 100 classes with 600 images each. The 600 images for each class were divided into 450 images to create a training set, 50 for a validation set and 100 for a test set. Note that the network is trained on all data from all classes available at present. First, it is trained on data from just five classes, then from all ten classes and so on, until finally, it is trained from data from all 100 classes simultaneously.
After each increment, the network was trained for 200 epochs, for a total of 4,000 epochs for all 20 increments. We used a learning-rate schedule that resets at the start of each increment. For the first 60 epochs of each increment, the learning rate was set to 0.1, then to 0.02 for the next 60 epochs, then 0.004 for the next 40 epochs and to 0.0008 for the last 40 epochs; we used the initial learning rate and learning-rate schedule reported in ref. 53 . During the 200 epochs of training for each increment, we kept track of the network with the best accuracy on the validation set. To prevent overfitting, at the start of each new increment, we reset the weights of the network to the weights of the best-performing (on the validation set) network found during the previous increment; this is equivalent to early stopping for each different increment.
We used an 18-layer deep residual network 38 for all experiments on class-incremental CIFAR-100. The network architecture is described in detail in Extended Data Table 1 . The weights of convolutional and linear layers were initialized using Kaiming initialization 54 , the weights for the batch-norm layers were initialized to one and all of the bias terms in the network were initialized to zero. Each time five new classes were made available to the network, five more output units were added to the final layer of the network. The weights and biases of these output units were initialized using the same initialization scheme. The weights of the network were optimized using SGD with a momentum of 0.9, a weight decay of 0.0005 and a mini-batch size of 90.
We used several steps of data preprocessing before the images were presented to the network. First, the value of all the pixels in each image was rescaled between 0 and 1 through division by 255. Then, each pixel in each channel was centred and rescaled by the average and standard deviation of the pixel values of each channel, respectively. Finally, we applied three random data transformations to each image before feeding it to the network: randomly horizontally flip the image with a probability of 0.5, randomly crop the image by padding the image with 4 pixels on each side and randomly cropping to the original size, and randomly rotate the image between 0 and 15°. The first two steps of preprocessing were applied to the training, validation and test sets, but the random transformations were only applied to the images in the training set.
We tested several hyperparameters to ensure the best performance for each different algorithm with our specific architecture. For the base system, we tested values for the weight decay parameter in {0.005, 0.0005, 0.00005}. A weight-decay value of 0.0005 resulted in the best performance in terms of area under the curve for accuracy on the test set over the 20 increments. For Shrink and Perturb, we used the weight-decay value of the base system and tested values for the standard deviation of the Gaussian noise in {10 −4 , 10 −5 , 10 −6 }; 10 −5 resulted in the best performance. For continual backpropagation, we tested values for the maturity threshold in {1,000, 10,000} and for the replacement rate in {10 −4 , 10 −5 , 10 −6 } using the contribution utility described in equation ( 1 ). A maturity threshold of 1,000 and a replacement rate of 10 −5 resulted in the best performance. Finally, for the head-resetting baseline, in Extended Data Fig. 1a , we used the same hyperparameters as for the base system, but the output layer was reinitialized at the start of each increment.
In Fig. 2d , we plot the stable rank of the representation in the penultimate layer of the network and the percentage of dead units in the full network. For a matrix \({\boldsymbol{\Phi }}\in {{\mathbb{R}}}^{n\times m}\) with singular values σ k sorted in descending order for k = 1, 2,…, q and q = max( n , m ), the stable rank 55 is \(\min \left\{k:\frac{{\Sigma }_{i}^{k}{\sigma }_{i}}{{\Sigma }_{j}^{q}{\sigma }_{j}} > 0.99\right\}\) .
For reference, we also implemented a network with the same hyperparameters as the base system but that was reinitialized at the beginning of each increment. Figure 2b shows the performance of each algorithm relative to the performance of the reinitialized network. For completeness, Extended Data Fig. 1a shows the test accuracy of each algorithm in each different increment. The final accuracy of continual backpropagation on all 100 classes was 76.13%, whereas Extended Data Fig. 1b shows the performance of continual backpropagation for different replacement rates with a maturity threshold of 1,000. For all algorithms that we tested, there was no correlation between when a class was presented and the accuracy of that class, implying that the temporal order of classes did not affect performance.
Robust loss of plasticity in permuted MNIST
We now use a computationally cheap problem based on the MNIST dataset 56 to test the generality of loss of plasticity across various conditions. MNIST is one of the most common supervised-learning datasets used in deep learning. It consists of 60,000, 28 × 28, greyscale images of handwritten digits from 0 to 9, together with their correct labels. For example, the left image in Extended Data Fig. 3a shows an image that is labelled by the digit 7. The smaller number of classes and the simpler images enable much smaller networks to perform well on this dataset than are needed on ImageNet or CIFAR-100. The smaller networks in turn mean that much less computation is needed to perform the experiments and thus experiments can be performed in greater quantities and under a variety of different conditions, enabling us to perform deeper and more extensive studies of plasticity.
We created a continual supervised-learning problem using permuted MNIST datasets 57 , 58 . An individual permuted MNIST dataset is created by permuting the pixels in the original MNIST dataset. The right image in Extended Data Fig. 3a is an example of such a permuted image. Given a way of permuting, all 60,000 images are permuted in the same way to produce the new permuted MNIST dataset. Furthermore, we normalized pixel values between 0 and 1 by dividing by 255.
By repeatedly randomly selecting from the approximately 10 1930 possible permutations, we created a sequence of 800 permuted MNIST datasets and supervised-learning tasks. For each task, we presented each of its 60,000 images one by one in random order to the learning network. Then we moved to the next permuted MNIST task and repeated the whole procedure, and so on for up to 800 tasks. No indication was given to the network at the time of task switching. With the pixels being permuted in a completely unrelated way, we might expect classification performance to fall substantially at the time of each task switch. Nevertheless, across tasks, there could be some savings, some improvement in speed of learning or, alternatively, there could be loss of plasticity—loss of the ability to learn across tasks. The network was trained on a single pass through the data and there were no mini-batches. We call this problem Online Permuted MNIST.
We applied feed-forward neural networks with three hidden layers to Online Permuted MNIST. We did not use convolutional layers, as they could not be helpful on the permuted problem because the spatial information is lost; in MNIST, convolutional layers are often not used even on the standard, non-permuted problem. For each example, the network estimated the probabilities of each of the tem classes, compared them to the correct label and performed SGD on the cross-entropy loss. As a measure of online performance, we recorded the percentage of times the network correctly classified each of the 60,000 images in the task. We plot this per-task performance measure versus task number in Extended Data Fig. 3b . The weights were initialized according to a Kaiming distribution.
The left panel of Extended Data Fig. 3b shows the progression of online performance across tasks for a network with 2,000 units per layer and various values of the step-size parameter. Note that that performance first increased across tasks, then began falling steadily across all subsequent tasks. This drop in performance means that the network is slowly losing the ability to learn from new tasks. This loss of plasticity is consistent with the loss of plasticity that we observed in ImageNet and CIFAR-100.
Next, we varied the network size. Instead of 2,000 units per layer, we tried 100, 1,000 and 10,000 units per layer. We ran this experiment for only 150 tasks, primarily because the largest network took much longer to run. The performances at good step sizes for each network size are shown in the middle panel of Extended Data Fig. 3b . Loss of plasticity with continued training is most pronounced at the smaller network sizes, but even the largest networks show some loss of plasticity.
Next, we studied the effect of the rate at which the task changed. Going back to the original network with 2,000-unit layers, instead of changing the permutation after each 60,000 examples, we now changed it after each 10,000, 100,000 or 1 million examples and ran for 48 million examples in total no matter how often the task changed. The examples in these cases were selected randomly with replacement for each task. As a performance measure of the network on a task, we used the percentage correct over all of the images in the task. The progression of performance is shown in the right panel in Extended Data Fig. 3b . Again, performance fell across tasks, even if the change was very infrequent. Altogether, these results show that the phenomenon of loss of plasticity robustly arises in this form of backpropagation. Loss of plasticity happens for a wide range of step sizes, rates of distribution change and for both underparameterized and overparameterized networks.
Loss of plasticity with different activations in the Slowly-Changing Regression problem
There remains the issue of the network’s activation function. In our experiments so far, we have used ReLU, the most popular choice at present, but there are several other possibilities. For these experiments, we switched to an even smaller, more idealized problem. Slowly-Changing Regression is a computationally inexpensive problem in which we can run a single experiment on a CPU core in 15 min, allowing us to perform extensive studies. As its name suggests, this problem is a regression problem—meaning that the labels are real numbers, with a squared loss, rather than nominal values with a cross-entropy loss—and the non-stationarity is slow and continual rather than abrupt, as in a switch from one task to another. In Slowly-Changing Regression, we study loss of plasticity for networks with six popular activation functions: sigmoid, tanh, ELU 59 , leaky ReLU 60 , ReLU 61 and Swish 62 .
In Slowly-Changing Regression, the learner receives a sequence of examples. The input for each example is a binary vector of size m + 1. The input has f slow-changing bits, m − f random bits and then one constant bit. The first f bits in the input vector change slowly. After every T examples, one of the first f bits is chosen uniformly at random and its value is flipped. These first f bits remain fixed for the next T examples. The parameter T allows us to control the rate at which the input distribution changes. The next m − f bits are randomly sampled for each example. Last, the ( m + 1)th bit is a bias term with a constant value of one.
The target output is generated by running the input vector through a neural network, which is set at the start of the experiment and kept fixed. As this network generates the target output and represents the desired solution, we call it the target network. The weights of the target networks are randomly chosen to be +1 or −1. The target network has one hidden layer with the linear threshold unit (LTU) activation. The value of the i th LTU is one if the input is above a threshold θ i and 0 otherwise. The threshold θ i is set to be equal to ( m + 1) × β − S i , in which β ∈ [0, 1] and S i is the number of input weights with negative value 63 . The details of the input and target function in the Slowly-Changing Regression problem are also described in Extended Data Fig. 2a .
The details of the specific instance of the Slowly-Changing Regression problem we use in this paper and the learning network used to predict its output are listed in Extended Data Table 4 . Note that the target network is more complex than the learning network, as the target network is wider, with 100 hidden units, whereas the learner has just five hidden units. Thus, because the input distribution changes every T example and the target function is more complex than what the learner can represent, there is a need to track the best approximation.
We applied learning networks with different activation functions to the Slowly-Changing Regression. The learner used the backpropagation algorithm to learn the weights of the network. We used a uniform Kaiming distribution 54 to initialize the weights of the learning network. The distribution is U (− b , b ) with bound, \(b={\rm{g}}{\rm{a}}{\rm{i}}{\rm{n}}\times \sqrt{\frac{3}{{\rm{n}}{\rm{u}}{\rm{m}}{\rm{\_}}{\rm{i}}{\rm{n}}{\rm{p}}{\rm{u}}{\rm{t}}{\rm{s}}}}\) , in which gain is chosen such that the magnitude of inputs does not change across layers. For tanh, sigmoid, ReLU and leaky ReLU, the gain is 5/3, 1, \(\sqrt{2}\) and \(\sqrt{2/(1+{\alpha }^{2})}\) , respectively. For ELU and Swish, we used \({\rm{gain}}=\sqrt{2}\) , as was done in the original papers 59 , 62 .
We ran the experiment on the Slowly-Changing Regression problem for 3 million examples. For each activation and value of step size, we performed 100 independent runs. First, we generated 100 sequences of examples (input–output pairs) for the 100 runs. Then these 100 sequences of examples were used for experiments with all activations and values of the step-size parameter. We used the same sequence of examples to control the randomness in the data stream across activations and step sizes.
The results of the experiments are shown in Extended Data Fig. 2b . We measured the squared error, that is, the square of the difference between the true target and the prediction made by the learning network. In Extended Data Fig. 2b , the squared error is presented in bins of 40,000 examples. This means that the first data point is the average squared error on the first 40,000 examples, the next is the average squared error on the next 40,000 examples and so on. The shaded region in the figure shows the standard error of the binned error.
Extended Data Fig. 2b shows that, in Slowly-Changing Regression, after performing well initially, the error increases for all step sizes and activations. For some activations such as ReLU and tanh, loss of plasticity is severe, and the error increases to the level of the linear baseline. Although for other activations such as ELU loss of plasticity is less severe, there is still a notable loss of plasticity. These results mean that loss of plasticity is not resolved by using commonly used activations. The results in this section complement the results in the rest of the article and add to the generality of loss of plasticity in deep learning.
Understanding loss of plasticity
We now turn our attention to understanding why backpropagation loses plasticity in continual-learning problems. The only difference in the learner over time is the network weights. In the beginning, the weights were small random numbers, as they were sampled from the initial distribution; however, after learning some tasks, the weights became optimized for the most recent task. Thus, the starting weights for the next task are qualitatively different from those for the first task. As this difference in the weights is the only difference in the learning algorithm over time, the initial weight distribution must have some unique properties that make backpropagation plastic in the beginning. The initial random distribution might have many properties that enable plasticity, such as the diversity of units, non-saturated units, small weight magnitude etc.
As we now demonstrate, many advantages of the initial distribution are lost concurrently with loss of plasticity. The loss of each of these advantages partially explains the degradation in performance that we have observed. We then provide arguments for how the loss of these advantages could contribute to loss of plasticity and measures that quantify the prevalence of each phenomenon. We provide an in-depth study of the Online Permuted MNIST problem that will serve as motivation for several solution methods that could mitigate loss of plasticity.
The first noticeable phenomenon that occurs concurrently with the loss of plasticity is the continual increase in the fraction of constant units. When a unit becomes constant, the gradients flowing back from the unit become zero or very close to zero. Zero gradients mean that the weights coming into the unit do not change, which means that this unit loses all of its plasticity. In the case of ReLU activations, this occurs when the output of the activations is zero for all examples of the task; such units are often said to be dead 64 , 65 . In the case of the sigmoidal activation functions, this phenomenon occurs when the output of a unit is too close to either of the extreme values of the activation function; such units are often said to be saturated 66 , 67 .
To measure the number of dead units in a network with ReLU activation, we count the number of units with a value of zero for all examples in a random sample of 2,000 images at the beginning of each new task. An analogous measure in the case of sigmoidal activations is the number of units that are ϵ away from either of the extreme values of the function for some small positive ϵ (ref. 68 ). We only focus on ReLU networks in this section.
In our experiments on the Online Permuted MNIST problem, the deterioration of online performance is accompanied by a large increase in the number of dead units (left panel of Extended Data Fig. 3c ). For the step size of 0.01, up to 25% of units die after 800 tasks. In the permuted MNIST problem, in which all inputs are positive because they are normalized between 0 and 1, once a unit in the first layer dies, it stays dead forever. Thus, an increase in dead units directly decreases the total capacity of the network. In the next section, we will see that methods that stop the units from dying can substantially reduce loss of plasticity. This further supports the idea that the increase in dead units is one of the causes of loss of plasticity in backpropagation.
Another phenomenon that occurs with loss of plasticity is the steady growth of the network’s average weight magnitude. We measure the average magnitude of the weights by adding up their absolute values and dividing by the total number of weights in the network. In the permuted MNIST experiment, the degradation of online classification accuracy of backpropagation observed in Extended Data Fig. 3b is associated with an increase in the average magnitude of the weights (centre panel of Extended Data Fig. 3c ). The growth of the magnitude of the weights of the network can represent a problem because large weight magnitudes are often associated with slower learning. The weights of a neural network are directly linked to the condition number of the Hessian matrix in the second-order Taylor approximation of the loss function. The condition number of the Hessian is known to affect the speed of convergence of SGD algorithms (see ref. 69 for an illustration of this phenomenon in convex optimization). Consequently, the growth in the magnitude of the weights could lead to an ill-conditioned Hessian matrix, resulting in a slower convergence.
The last phenomenon that occurs with the loss of plasticity is the drop in the effective rank of the representation. Similar to the rank of a matrix, which represents the number of linearly independent dimensions, the effective rank takes into consideration how each dimension influences the transformation induced by a matrix 70 . A high effective rank indicates that most of the dimensions of the matrix contribute similarly to the transformation induced by the matrix. On the other hand, a low effective rank corresponds to most dimensions having no notable effect on the transformation, implying that the information in most of the dimensions is close to being redundant.
Formally, consider a matrix \(\Phi \in {{\mathbb{R}}}^{n\times m}\) with singular values σ k for k = 1, 2,…, q , and q = max( n , m ). Let p k = σ k / ∥ σ ∥ 1 , in which σ is the vector containing all the singular values and ∥ ⋅ ∥ 1 is the ℓ 1 norm. The effective rank of matrix Φ , or erank( Φ ), is defined as
Note that the effective rank is a continuous measure that ranges between one and the rank of matrix Φ .
In the case of neural networks, the effective rank of a hidden layer measures the number of units that can produce the output of the layer. If a hidden layer has a low effective rank, then a small number of units can produce the output of the layer, meaning that many of the units in the hidden layer are not providing any useful information. We approximate the effective rank on a random sample of 2,000 examples before training on each task.
In our experiments, loss of plasticity is accompanied by a decrease in the average effective rank of the network (right panel of Extended Data Fig. 3c ). This phenomenon in itself is not necessarily a problem. After all, it has been shown that gradient-based optimization seems to favour low-rank solutions through implicit regularization of the loss function or implicit minimization of the rank itself 71 , 72 . However, a low-rank solution might be a bad starting point for learning from new observations because most of the hidden units provide little to no information.
The decrease in effective rank could explain the loss of plasticity in our experiments in the following way. After each task, the learning algorithm finds a low-rank solution for the current task, which then serves as the initialization for the next task. As the process continues, the effective rank of the representation layer keeps decreasing after each task, limiting the number of solutions that the network can represent immediately at the start of each new task.
In this section, we looked deeper at the networks that lost plasticity in the Online Permuted MNIST problem. We noted that the only difference in the learning algorithm over time is the weights of the network, which means that the initial weight distribution has some properties that allowed the learning algorithm to be plastic in the beginning. And as learning progressed, the weights of the network moved away from the initial distribution and the algorithm started to lose plasticity. We found that loss of plasticity is correlated with an increase in weight magnitude, a decrease in the effective rank of the representation and an increase in the fraction of dead units. Each of these correlates partially explains loss of plasticity faced by backpropagation.
![](http://sektorel.online/777/templates/cheerup/res/banner1.gif)
Existing deep-learning methods for mitigating loss of plasticity
We now investigate several existing methods and test how they affect loss of plasticity. We study five existing methods: L2 regularization 73 , Dropout 74 , online normalization 75 , Shrink and Perturb 11 and Adam 43 . We chose L2 regularization, Dropout, normalization and Adam because these methods are commonly used in deep-learning practice. Although Shrink and Perturb is not a commonly used method, we chose it because it reduces the failure of pretraining, a problem that is an instance of loss of plasticity. To assess if these methods can mitigate loss of plasticity, we tested them on the Online Permuted MNIST problem using the same network architecture we used in the previous section, ‘Understanding loss of plasticity’. Similar to the previous section, we measure the online classification accuracy on all 60,000 examples of the task. All the algorithms used a step size of 0.003, which was the best-performing step size for backpropagation in the left panel of Extended Data Fig. 3b . We also use the three correlates of loss of plasticity found in the previous section to get a deeper understanding of the performance of these methods.
An intuitive way to address loss of plasticity is to use weight regularization, as loss of plasticity is associated with a growth of weight magnitudes, shown in the previous section. We used L2 regularization, which adds a penalty to the loss function proportional to the ℓ 2 norm of the weights of the network. The L2 regularization penalty incentivizes SGD to find solutions that have a low weight magnitude. This introduces a hyperparameter λ that modulates the contribution of the penalty term.
The purple line in the left panel of Extended Data Fig. 4a shows the performance of L2 regularization on the Online Permuted MNIST problem. The purple lines in the other panels of Extended Data Fig. 4a show the evolution of the three correlates of loss of plasticity with L2 regularization. For L2 regularization, the weight magnitude does not continually increase. Moreover, as expected, the non-increasing weight magnitude is associated with lower loss of plasticity. However, L2 regularization does not fully mitigate loss of plasticity. The other two correlates for loss of plasticity explain this, as the percentage of dead units kept increasing and the effective rank kept decreasing. Finally, Extended Data Fig. 4b shows the performance of L2 regularization for different values of λ . The regularization parameter λ controlled the peak of the performance and how quickly it decreased.
A method related to weight regularization is Shrink and Perturb 11 . As the name suggests, Shrink and Perturb performs two operations; it shrinks all the weights and then adds random Gaussian noise to these weights. The introduction of noise introduces another hyperparameter, the standard deviation of the noise. Owing to the shrinking part of Shrink and Perturb, the algorithm favours solutions with smaller average weight magnitude than backpropagation. Moreover, the added noise prevents units from dying because it adds a non-zero probability that a dead unit will become active again. If Shrink and Perturb mitigates these correlates to loss of plasticity, it could reduce loss of plasticity.
The performance of Shrink and Perturb is shown in orange in Extended Data Fig. 4 . Similar to L2 regularization, Shrink and Perturb stops the weight magnitude from continually increasing. Moreover, it also reduces the percentage of dead units. However, it has a lower effective rank than backpropagation, but still higher than that of L2 regularization. Not only does Shrink and Perturb have a lower loss of plasticity than backpropagation but it almost completely mitigates loss of plasticity in Online Permuted MNIST. However, Shrink and Perturb was sensitive to the standard deviation of the noise. If the noise was too high, loss of plasticity was much more severe, and if it was too low, it did not have any effect.
An important technique in modern deep learning is called Dropout 74 . Dropout randomly sets each hidden unit to zero with a small probability, which is a hyperparameter of the algorithm. The performance of Dropout is shown in pink in Extended Data Fig. 4 .
Dropout showed similar measures of percentage of dead units, weight magnitude and effective rank as backpropagation, but, surprisingly, showed higher loss of plasticity. The poor performance of Dropout is not explained by our three correlates of loss of plasticity, which means that there are other possible causes of loss of plasticity. A thorough investigation of Dropout is beyond the scope of this paper, though it would be an interesting direction for future work. We found that a higher Dropout probability corresponded to a faster and sharper drop in performance. Dropout with probability of 0.03 performed the best and its performance was almost identical to that of backpropagation. However, Extended Data Fig. 4a shows the performance for a Dropout probability of 0.1 because it is more representative of the values used in practice.
Another commonly used technique in deep learning is batch normalization 76 . In batch normalization, the output of each hidden layer is normalized and rescaled using statistics computed from each mini-batch of data. We decided to include batch normalization in this investigation because it is a popular technique often used in practice. Because batch normalization is not amenable to the online setting used in the Online Permuted MNIST problem, we used online normalization 77 instead, an online variant of batch normalization. Online normalization introduces two hyperparameters used for the incremental estimation of the statistics in the normalization steps.
The performance of online normalization is shown in green in Extended Data Fig. 4 . Online normalization had fewer dead units and a higher effective rank than backpropagation in the earlier tasks, but both measures deteriorated over time. In the later tasks, the network trained using online normalization has a higher percentage of dead units and a lower effective rank than the network trained using backpropagation. The online classification accuracy is consistent with these results. Initially, it has better classification accuracy, but later, its classification accuracy becomes lower than that of backpropagation. For online normalization, the hyperparameters changed when the performance of the method peaked, and it also slightly changed how fast it got to its peak performance.
No assessment of alternative methods can be complete without Adam 43 , as it is considered one of the most useful tools in modern deep learning. The Adam optimizer is a variant of SGD that uses an estimate of the first moment of the gradient scaled inversely by an estimate of the second moment of the gradient to update the weights instead of directly using the gradient. Because of its widespread use and success in both supervised and reinforcement learning, we decided to include Adam in this investigation to see how it would affect the plasticity of deep neural networks. Adam has two hyperparameters that are used for computing the moving averages of the first and second moments of the gradient. We used the default values of these hyperparameters proposed in the original paper and tuned the step-size parameter.
The performance of Adam is shown in cyan in Extended Data Fig. 4 . Adam’s loss of plasticity can be categorized as catastrophic, as it reduces substantially. Consistent with our previous results, Adam scores poorly in the three measures corresponding to the correlates of loss of plasticity. Adam had an early increase in the percentage of dead units that plateaus at around 60%, similar weight magnitude as backpropagation and a large drop in the effective rank early during training. We also tested Adam with different activation functions on the Slowly-Changing Regression and found that loss of plasticity with Adam is usually worse than with SGD.
Many of the standard methods substantially worsened loss of plasticity. The effect of Adam on the plasticity of the networks was particularly notable. Networks trained with Adam quickly lost almost all of their diversity, as measured by the effective rank, and gained a large percentage of dead units. This marked loss of plasticity of Adam is an important result for deep reinforcement learning, for which Adam is the default optimizer 78 , and reinforcement learning is inherently continual owing to the ever-changing policy. Similar to Adam, other commonly used methods such as Dropout and normalization worsened loss of plasticity. Normalization had better performance in the beginning, but later it had a sharper drop in performance than backpropagation. In the experiment, Dropout simply made the performance worse. We saw that the higher the Dropout probability, the larger the loss of plasticity. These results mean that some of the most successful tools in deep learning do not work well in continual learning, and we need to focus on directly developing tools for continual learning.
We did find some success in maintaining plasticity in deep neural networks. L2 regularization and Shrink and Perturb reduce loss of plasticity. Shrink and Perturb is particularly effective, as it almost entirely mitigates loss of plasticity. However, both Shrink and Perturb and L2 regularization are slightly sensitive to hyperparameter values. Both methods only reduce loss of plasticity for a small range of hyperparameters, whereas for other hyperparameter values, they make loss of plasticity worse. This sensitivity to hyperparameters can limit the application of these methods to continual learning. Furthermore, Shrink and Perturb does not fully resolve the three correlates of loss of plasticity, it has a lower effective rank than backpropagation and it still has a high fraction of dead units.
We also applied continual backpropagation on Online Permuted MNIST. The replacement rate is the main hyperparameter in continual backpropagation, as it controls how rapidly units are reinitialized in the network. For example, a replacement rate of 10 −6 for our network with 2,000 hidden units in each layer would mean replacing one unit in each layer after every 500 examples.
Blue lines in Extended Data Fig. 4 show the performance of continual backpropagation. It has a non-degrading performance and is stable for a wide range of replacement rates. Continual backpropagation also mitigates all three correlates of loss of plasticity. It has almost no dead units, stops the network weights from growing and maintains a high effective rank across tasks. All algorithms that maintain a low weight magnitude also reduced loss of plasticity. This supports our claim that low weight magnitudes are important for maintaining plasticity. The algorithms that maintain low weight magnitudes were continual backpropagation, L2 regularization and Shrink and Perturb. Shrink and Perturb and continual backpropagation have an extra advantage over L2 regularization: they inject randomness into the network. This injection of randomness leads to a higher effective rank and lower number of dead units, which leads to both of these algorithms performing better than L2 regularization. However, continual backpropagation injects randomness selectively, effectively removing all dead units from the network and leading to a higher effective rank. This smaller number of dead units and a higher effective rank explains the better performance of continual backpropagation.
Details and further analysis in reinforcement learning
The experiments presented in the main text were conducted using the Ant-v3 environment from OpenAI Gym 79 . We changed the coefficient of friction by sampling it log-uniformly from the range [0.02, 2.00], using a logarithm with base 10. The coefficient of friction changed at the first episode boundary after 2 million time steps had passed since the last change. We also tested Shrink and Perturb on this problem and found that it did not provide a marked performance improvement over L2 regularization. Two separate networks were used for the policy and the value function, and both had two hidden layers with 256 units. These networks were trained using Adam alongside PPO to update the weights in the network. See Extended Data Table 5 for the values of the other hyperparameters. In all of the plots showing results of reinforcement-learning experiments, the shaded region represents the 95% bootstrapped confidence 80 .
The reward signal in the ant problem consists of four components. The main component rewards the agent for forward movement. It is proportional to the distance moved by the ant in the positive x direction since the last time step. The second component has a value of 1 at each time step. The third component penalizes the ant for taking large actions. This component is proportional to the square of the magnitude of the action. Finally, the last component penalizes the agent for large external contact forces. It is proportional to the sum of external forces (clipped in a range). The reward signal at each time step is the sum of these four components.
We also evaluated PPO and its variants in two more environments: Hopper-v3 and Walker-v3. The results for these experiments are presented in Extended Data Fig. 5a . The results mirrored those from Ant-v3; standard PPO suffered from a notable degradation in performance, in which its performance decreased substantially. However, this time, L2 regularization did not fix the issue in all cases; there was some performance degradation with L2 in Walker-v3. PPO, with continual backpropagation and L2 regularization, completely fixed the issue in all environments. Note that the only difference between our experiments and what is typically done in the literature is that we run the experiments for longer. Typically, these experiments are only done for 3 million steps, but we ran these experiments for up to 100 million steps.
PPO with L2 regularization only avoided degradation for a relatively large value of weight decay, 10 −3 . This extreme regularization stops the agent from finding better policies and stays stuck at a suboptimal policy. There was large performance degradation for smaller values of weight decay, and for larger values, the performance was always low. When we used continual backpropagation and L2 regularization together, we could use smaller values of weight decay. All the results for PPO with continual backpropagation and L2 regularization have a weight decay of 10 −4 , a replacement rate of 10 −4 and a maturity threshold of 10 4 . We found that the performance of PPO with continual backpropagation and L2 regularization was sensitive to the replacement rate but not to the maturity threshold and weight decay.
PPO uses the Adam optimizer, which keeps running estimates of the gradient and the squared of the gradient. These estimates require two further parameters, called β 1 and β 2 . The standard values of β 1 and β 2 are 0.9 and 0.999, respectively, which we refer to as standard Adam. Lyle et al. 24 showed that the standard values of β 1 and β 2 cause a large loss of plasticity. This happens because of the mismatch in β 1 and β 2 . A sudden large gradient can cause a very large update, as a large value of β 2 means that the running estimate for the square of the gradient, which is used in the denominator, is updated much more slowly than the running estimate for the gradient, which is the numerator. This loss of plasticity in Adam can be reduced by setting β 1 equal to β 2 . In our experiments, we set β 1 and β 2 to 0.99 and refer to it as tuned Adam/PPO. In Extended Data Fig. 5c , we measure the largest total weight change in the network during a single update cycle for bins of 1 million steps. The first point in the plots shows the largest weight change in the first 1 million steps. The second point shows the largest weight change in the second 1 second steps and so on. The figure shows that standard Adam consistently causes very large updates to the weights, which can destabilize learning, whereas tuned Adam with β 1 = β 2 = 0.99 has substantially smaller updates, which leads to more stable learning. In all of our experiments, all algorithms other than the standard PPO used the tuned parameters for Adam ( β 1 = β 2 = 0.99). The failure of standard Adam with PPO is similar to the failure of standard Adam in permuted MNIST.
In our next experiment, we perform a preliminary comparison with ReDo 25 . ReDo is another selective reinitialization method that builds on continual backpropagation but uses a different measure of utility and strategy for reinitializing. We tested ReDo on Ant-v3, the hardest of the three environments. ReDo requires two parameters: a threshold and a reinitialization period. We tested ReDo for all combinations of thresholds in {0.01, 0.03, 0.1} and reinitialization periods in {10, 10 2 , 10 3 , 10 4 , 10 5 }; a threshold of 0.1 with a reinitialization period of 10 2 performed the best. The performance of PPO with ReDo is plotted in Extended Data Fig. 5b . ReDo and continual backpropagation were used with weight decay of 10 −4 and β 1 and β 2 of 0.99. The figure shows that PPO with ReDo and L2 regularization performs much better than standard PPO. However, it still suffers from performance degradation and its performance is worse than PPO with L2 regularization. Note that this is only a preliminary comparison; we leave a full comparison and analysis of both methods for future work.
The performance drop of PPO in stationary environments is a nuanced phenomenon. Loss of plasticity and forgetting are both responsible for the observed degradation in performance. The degradation in performance implies that the agent forgot the good policy it had once learned, whereas the inability of the agent to relearn a good policy means it lost plasticity.
Loss of plasticity expresses itself in various forms in deep reinforcement learning. Some work found that deep reinforcement learning systems can lose their generalization abilities in the presence of non-stationarities 81 . A reduction in the effective rank, similar to the rank reduction in CIFAR-100, has been observed in some deep reinforcement-learning algorithms 82 . Nikishin et al. 18 showed that many reinforcement-learning systems perform better if their network is occasionally reset to its naive initial state, retaining only the replay buffer. This is because the learning networks became worse than a reinitialized network at learning from new data. Recent work has improved performance in many reinforcement-learning problems by applying plasticity-preserving methods 25 , 83 , 84 , 85 , 86 , 87 . These works focused on deep reinforcement learning systems that use large replay buffers. Our work complements this line of research as we studied systems based on PPO, which has much smaller replay buffers. Loss of plasticity is most relevant for systems that use small or no replay buffers, as large buffers can hide the effect of new data. Overcoming loss of plasticity is an important step towards deep reinforcement-learning systems that can learn from an online data stream.
Extended discussion
There are two main goals in continual learning: maintaining stability and maintaining plasticity 88 , 89 , 90 , 91 . Maintaining stability is concerned with memorizing useful information and maintaining plasticity is about finding new useful information when the data distribution changes. Current deep-learning methods struggle to maintain stability as they tend to forget previously learned information 28 , 29 . Many papers have been dedicated to maintaining stability in deep continual learning 30 , 92 , 93 , 94 , 95 , 96 , 97 . We focused on continually finding useful information, not on remembering useful information. Our work on loss of plasticity is different but complementary to the work on maintaining stability. Continual backpropagation in its current form does not tackle the forgetting problem. Its current utility measure only considers the importance of units for current data. One idea to tackle forgetting is to use a long-term measure of utility that remembers which units were useful in the past. Developing methods that maintain both stability and plasticity is an important direction for future work.
There are many desirable properties for an efficient continual-learning system 98 , 99 . It should be able to keep learning new things, control what it remembers and forgets, have good computational and memory efficiency and use previous knowledge to speed up learning on new data. The choice of the benchmark affects which property is being focused on. Most benchmarks and evaluations in our paper only focused on plasticity but not on other aspects, such as forgetting and speed of learning. For example, in Continual ImageNet, previous tasks are rarely repeated, which makes it effective for studying plasticity but not forgetting. In permuted MNIST, consecutive tasks are largely independent, which makes it suitable for studying plasticity in isolation. However, this independence means that previous knowledge cannot substantially speed up learning on new tasks. On the other hand, in class-incremental CIFAR-100, previous knowledge can substantially speed up learning of new classes. Overcoming loss of plasticity is an important, but still the first, step towards the goal of fast learning on future data 100 , 101 , 102 . Once we have networks that maintain plasticity, we can develop methods that use previous knowledge to speed up learning on future data.
Loss of plasticity is a critical factor when learning continues for many tasks, but it might be less important if learning happens for a small number of tasks. Usually, the learning system can take advantage of previous learning in the first few tasks. For example, in class-incremental CIFAR-100 (Fig. 2 ), the base deep-learning systems performed better than the network trained from scratch for up to 40 classes. This result is consistent with deep-learning applications in which the learning system is first trained on a large dataset and then fine-tuned on a smaller, more relevant dataset. Plasticity-preserving methods such as continual backpropagation may still improve performance in such applications based on fine-turning, but we do not expect that improvement to be large, as learning happens only for a small number of tasks. We have observed that deep-learning systems gradually lose plasticity, and this effect accumulates over tasks. Loss of plasticity becomes an important factor when learning continues for a large number of tasks; in class-incremental CIFAR-100, the performance of the base deep-learning system was much worse after 100 classes.
We have made notable progress in understanding loss of plasticity. However, it remains unclear which specific properties of initialization with small random numbers are important for maintaining plasticity. Recent work 103 , 104 has made exciting progress in this direction and it remains an important avenue for future work. The type of loss of plasticity studied in this article is largely because of the loss of the ability to optimize new objectives. This is different from the type of loss of plasticity in which the system can keep optimizing new objectives but lose the ability to generalize 11 , 12 . However, it is unclear if the two types of plasticity loss are fundamentally different or if the same mechanism can explain both phenomena. Future work that improves our understanding of plasticity and finds the underlying causes of both types of plasticity loss will be valuable to the community.
Continual backpropagation uses a utility measure to find and replace low-utility units. One limitation of continual backpropagation is that the utility measure is based on heuristics. Although it performs well, future work on more principled utility measures will improve the foundations of continual backpropagation. Our current utility measure is not a global measure of utility as it does not consider how a given unit affects the overall represented function. One possibility is to develop utility measures in which utility is propagated backwards from the loss function. The idea of utility in continual backpropagation is closely related to connection utility in the neural-network-pruning literature. Various papers 105 , 106 , 107 , 108 have proposed different measures of connection utility for the network-pruning problem. Adapting these utility measures to mitigate loss of plasticity is a promising direction for new algorithms and some recent work is already making progress in this direction 109 .
The idea of selective reinitialization is similar to the emerging idea of dynamic sparse training 110 , 111 , 112 . In dynamic sparse training, a sparse network is trained from scratch and connections between different units are generated and removed during training. Removing connections requires a measure of utility, and the initialization of new connections requires a generator similar to selective reinitialization. The main difference between dynamic sparse training and continual backpropagation is that dynamic sparse training operates on connections between units, whereas continual backpropagation operates on units. Consequently, the generator in dynamic sparse training must also decide which new connections to grow. Dynamic sparse training has achieved promising results in supervised and reinforcement-learning problems 113 , 114 , 115 , in which dynamic sparse networks achieve performance close to dense networks even at high sparsity levels. Dynamic sparse training is a promising idea that can be useful to maintain plasticity.
The idea of adding new units to neural networks is present in the continual-learning literature 92 , 116 , 117 . This idea is usually manifested in algorithms that dynamically increase the size of the network. For example, one method 117 expands the network by allocating a new subnetwork whenever there is a new task. These methods do not have an upper limit on memory requirements. Although these methods are related to the ideas in continual backpropagation, none are suitable for comparison, as continual backpropagation is designed for learning systems with finite memory, which are well suited for lifelong learning. And these methods would therefore require non-trivial modification to apply to our setting of finite memory.
Previous works on the importance of initialization have focused on finding the correct weight magnitude to initialize the weights. It has been shown that it is essential to initialize the weights so that the gradients do not become exponentially small in the initial layers of a network and the gradient is preserved across layers 54 , 66 . Furthermore, initialization with small weights is critical for sigmoid activations as they may saturate if the weights are too large 118 . Despite all this work on the importance of initialization, the fact that its benefits are only present initially but not continually has been overlooked, as these papers focused on cases in which learning has to be done just once, not continually.
Continual backpropagation selectively reinitializes low-utility units. One common strategy to deal with non-stationary data streams is reinitializing the network entirely. In the Online Permuted MNIST experiment, full reinitialization corresponds to a performance that stays at the level of the first point (Extended Data Fig. 4a ). In this case, continual backpropagation outperforms full reinitialization as it takes advantage of what it has previously learned to speed up learning on new data. In ImageNet experiments, the final performance of continual backpropagation is only slightly better than a fully reinitialized network (the first point for backpropagation in left panel of Fig. 1b ). However, Fig. 1 does not show how fast an algorithm reaches the final performance in each task. We observed that continual backpropagation achieves the best accuracy ten times faster than a fully reinitialized network on the 5,000th task of Continual ImageNet, ten epochs versus about 125 epochs. Furthermore, continual backpropagation could be combined with other methods that mitigate forgetting, which can further speed up learning on new data. In reinforcement learning, full reinitialization is only practical for systems with a large buffer. For systems that keep a small or no buffer, such as those we studied, full reinitialization will lead the agent to forget everything it has learned, and its performance will be down to the starting point.
Loss of plasticity might also be connected to the lottery ticket hypothesis 119 . The hypothesis states that randomly initialized networks contain subnetworks that can achieve performance close to that of the original network with a similar number of updates. These subnetworks are called winning tickets. We found that, in continual-learning problems, the effective rank of the representation at the beginning of tasks reduces over time. In a sense, the network obtained after training on several tasks has less randomness and diversity than the original random network. The reduced randomness might mean that the network has fewer winning tickets. And this reduced number of winning tickets might explain loss of plasticity. Our understanding of loss of plasticity could be deepened by fully exploring its connection with the lottery ticket hypothesis.
Some recent works have focused on quickly adapting to the changes in the data stream 120 , 121 , 122 . However, the problem settings in these papers were offline as they had two separate phases, one for learning and the other for evaluation. To use these methods online, they have to be pretrained on tasks that represent tasks that the learner will encounter during the online evaluation phase. This requirement of having access to representative tasks in the pretraining phase is not realistic for lifelong learning systems as the real world is non-stationary, and even the distribution of tasks can change over time. These methods are not comparable with those we studied in our work, as we studied fully online methods that do not require pretraining.
In this work, we found that methods that continually injected randomness while maintaining small weight magnitudes greatly reduced loss of plasticity. Many works have found that adding noise while training neural networks can improve training and testing performance. The main benefits of adding noise have been reported to be avoiding overfitting and improving training performance 123 , 124 , 125 . However, it can be tricky to inject noise without degrading performance in some cases 126 . In our case, when the data distribution is non-stationary, we found that continually injecting noise along with L2 regularization helps with maintaining plasticity in neural networks.
Data availability
All of the datasets and simulation environments used in this work are publicly available. Other data needed to evaluate the conclusions in the article are present in the article or the extended data.
Code availability
The code is available at https://github.com/shibhansh/loss-of-plasticity .
Rumelhart, D. E., Hinton, G. E. & Williams, R. J. Learning representations by back-propagating errors. Nature 323 , 533–536 (1986).
Article ADS Google Scholar
OpenAI et al. GPT-4 technical report. Preprint at https://arxiv.org/abs/2303.08774 (2023).
Bahdanau, D., Cho, K. & Bengio, Y. Neural machine translation by jointly learning to align and translate. In Proc. 3rd International Conference on Learning Representations (eds Bengio, Y. & LeCun, Y.) (ICLR, 2015).
Jumper, J. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596 , 583–589 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Mnih, V. et al. Human-level control through deep reinforcement learning. Nature 518 , 529–533 (2015).
Article ADS CAS PubMed Google Scholar
Silver, D. et al. Mastering the game of Go with deep neural networks and tree search. Nature 529 , 484–489 (2016).
Moravčík, M. et al. DeepStack: expert-level artificial intelligence in heads-up no-limit poker. Science 356 , 508–513 (2017).
Article ADS MathSciNet PubMed Google Scholar
Wurman, P. R. et al. Outracing champion Gran Turismo drivers with deep reinforcement learning. Nature 602 , 223–228 (2022).
Andrychowicz, O. M. et al. Learning dexterous in-hand manipulation. Int. J. Robot. Res. 39 , 3–20 (2020).
Article Google Scholar
Kaufmann, E. et al. Champion-level drone racing using deep reinforcement learning. Nature 620 , 982–987 (2023).
Ash, J. & Adams, R. P. On warm-starting neural network training. Adv. Neural Inf. Process. Syst. 33 , 3884–3894 (2020).
Google Scholar
Berariu, T. et al. A study on the plasticity of neural networks. Preprint at https://arxiv.org/abs/2106.00042 (2021).
Ellis, A. W. & Lambon Ralph, M. A. Age of acquisition effects in adult lexical processing reflect loss of plasticity in maturing systems: insights from connectionist networks. J. Exp. Psychol. Learn. Mem. Cogn. 26 , 1103 (2000).
Article CAS PubMed Google Scholar
Zevin, J. D. & Seidenberg, M. S. Age of acquisition effects in word reading and other tasks. J. Mem. Lang. 47 , 1–29 (2002).
Bonin, P., Barry, C., Méot, A. & Chalard, M. The influence of age of acquisition in word reading and other tasks: a never ending story? J. Mem. Lang. 50 , 456–476 (2004).
Chaudhry, A., Dokania, P. K., Ajanthan, T. & Torr, P. H. Riemannian walk for incremental learning: understanding forgetting and intransigence. In Proc. 15th European Conference on Computer Vision (ECCV) 532–547 (Springer, 2018).
Achille, A., Rovere, M. & Soatto, S. Critical learning periods in deep networks. In Proc. 6th International Conference on Learning Representations (eds Murray, I., Ranzato, M. & Vinyals, O.) (ICLR, 2018).
Nikishin, E., Schwarzer, M., D’Oro, P., Bacon, P.-L. & Courville, A. The primacy bias in deep reinforcement learning. In Proc. 39th International Conference on Machine Learning 16828–16847 (PMLR, 2022).
Dohare, S. The Interplay of Search and Gradient Descent in Semi-stationary Learning Problems . Master’s thesis, Univ. Alberta (2020).
Rahman, P. Toward Generate-and-test Algorithms for Continual Feature Discovery . Master’s thesis, Univ. Alberta (2021).
Dohare, S., Sutton, R. S. & Mahmood, A. R. Continual backprop: stochastic gradient descent with persistent randomness. Preprint at https://arxiv.org/abs/2108.06325 (2021).
Lyle, C., Rowland, M. & Dabney, W. Understanding and preventing capacity loss in reinforcement learning. In Proc. 10th International Conference on Learning Representations (ICLR, 2022).
Abbas, Z., Zhao, R., Modayil, J., White, A. & Machado, M. C. Loss of plasticity in continual deep reinforcement learning. In Proc. 2nd Conference on Lifelong Learning Agents (PMLR, 2023).
Lyle, C. et al. Understanding plasticity in neural networks. In Proc. 40th International Conference on Machine Learning 23190–23211 (PMLR, 2023).
Sokar, G., Agarwal, R., Castro, P. S. & Evci, U. The dormant neuron phenomenon in deep reinforcement learning. In Proc. 40th International Conference on Machine Learning 32145–32168 (PMLR, 2023).
Dohare, S., Hernandez-Garcia, J. F., Rahman, P., Mahmood, A. R. & Sutton, R. S. Maintaining plasticity in deep continual learning. Preprint at https://arxiv.org/abs/2306.13812 (2023).
Kumar, S., Marklund, H. & Van Roy, B. Maintaining plasticity in continual learning via regenerative regularization. In Proc. 3rd Conference on Lifelong Learning Agents (PMLR, 2024).
McCloskey, M. & Cohen, N. J. Catastrophic interference in connectionist networks: the sequential learning problem. Psychol. Learn. Motiv. 24 , 109–165 (1989).
French, R. M. Catastrophic forgetting in connectionist networks. Trends Cogn. Sci. 3 , 128–135 (1999).
Kirkpatrick, J. et al. Overcoming catastrophic forgetting in neural networks. Proc. Natl Acad. Sci. 114 , 3521–3526 (2017).
Article ADS MathSciNet CAS PubMed Google Scholar
Selfridge, O. G. Pandemonium: a paradigm for learning. In Mechanization of Thought Processes: Proceedings of a Symposium Held at the National Physical Laboratory 511–531 (Her Majesty’s Stationary Office, 1958).
Klopf, A. H. & Gose, E. An evolutionary pattern recognition network. IEEE Trans. Syst. Sci. Cybern. 5 , 247–250 (1969).
Holland, J. H. & Reitman, J. S. Cognitive systems based on adaptive algorithms. ACM SIGART Bull. 63 , 49–49 (1977).
Kaelbling, L. P. Learning in Embedded Systems (MIT Press, 1993).
Mahmood, A. R. & Sutton, R. S. Representation search through generate and test. In Proc. AAAI Workshop: Learning Rich Representations from Low-Level Sensors 16–21 (2013).
Deng, J. et al. ImageNet: a large-scale hierarchical image database. In Proc. 2009 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 248–255 (IEEE, 2009).
Krizhevsky, A., Sutskever, I. & Hinton, G. E. ImageNet classification with deep convolutional neural networks. Adv. Neural Inf. Process. Syst. 25 , 1097–1105 (2012).
He, K., Zhang, X., Ren, S. & Sun, J. Deep residual learning for image recognition. In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 770–778 (2016).
Rebuffi, S.-A., Kolesnikov, A., Sperl, G. & Lampert, C. H. iCaRL: incremental classifier and representation learning. In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2001–2010 (2017).
Schulman, J., Wolski, F., Dhariwal, P., Radford, A. & Klimov, O. Proximal policy optimization algorithms. Preprint at https://arxiv.org/abs/1707.06347 (2017).
OpenAI et al. Dota 2 with large scale deep reinforcement learning. Preprint at https://arxiv.org/abs/1912.06680 (2019).
Ouyang, L. et al. Training language models to follow instructions with human feedback. Adv. Neural Inf. Process. Syst. 36 , 27730–27744 (2022).
Kingma, D. P. & Ba, J. Adam: a method for stochastic optimization. In Proc. 3rd International Conference on Learning Representations (eds Bengio, Y. & LeCun, Y.) (ICLR, 2015).
Campbell, D. T. Blind variation and selective survival as a general strategy in knowledge-processes. Psychol. Rev. 67 , 380–400 (1960).
Thorndike, E. L. Animal Intelligence (Macmillan, 1911).
Dennett, D. C. Why the law of effect will not go away. J. Theory Soc. Behav. 5 , 169–187 (1975).
Holland, J. H. Adaptation in Natural and Artificial Systems (MIT Press, 1992).
Kashyap, R., Blaydon, C. & Fu, K. in Adaptive, Learning and Pattern Recognition Systems: Theory and Applications (eds Mendel, J. & Fu, K.) 329–355 (Elsevier, 1970).
Powell, M. J. D. Restart procedures for the conjugate gradient method. Math. Program. 12 , 241–254 (1977).
Article MathSciNet Google Scholar
Stanley, K. O. & Miikkulainen, R. Evolving neural networks through augmenting topologies. Evol. Comput. 10 , 99–127 (2002).
Article PubMed Google Scholar
Chrabaszcz, P., Loshchilov, I. & Hutter, F. A downsampled variant of ImageNet as an alternative to the CIFAR datasets. Preprint at https://arxiv.org/abs/1707.08819 (2017).
van de Ven, G. M., Tuytelaars, T. & Tolias, A. S. Three types of incremental learning. Nat. Mach. Intell. 4 , 1185–1197 (2022).
Weiaicunzai. pytorch-cifar100. GitHub https://github.com/weiaicunzai/pytorch-cifar100 (2022).
He, K., Zhang, X., Ren, S. & Sun, J. Delving deep into rectifiers: surpassing human-level performance on ImageNet classification. In Proc. IEEE International Conference on Computer Vision (ICCV) 1026–1034 (IEEE, 2015).
Yang, Y., Zhang, G., Xu, Z. & Katabi, D. Harnessing structures for value-based planning and reinforcement learning. In Proc. 7th International Conference on Learning Representations (ICLR, 2019).
Lecun, Y., Bottou, L., Bengio, Y. & Haffner, P. Gradient-based learning applied to document recognition. Proc. IEEE 86 , 2278–2324 (1998).
Goodfellow, I., Mirza, M., Xiao, D. & Aaron Courville, Y. B. An empirical investigation of catastrophic forgeting in gradient-based neural networks. In Proc. 2nd International Conference on Learning Representations (ICLR, 2014).
Zenke, F., Poole, B. & Ganguli, S. Continual learning through synaptic intelligence. In Proc. 34th International Conference on Machine Learning 3987–3995 (PMLR, 2017).
Clevert, D., Unterthiner, T. & Hochreiter, S. Fast and accurate deep network learning by exponential linear units (ELUs). In Proc. 4th International Conference on Learning Representations (ICLR, 2016).
Maas, A. L., Hannun, A. Y. & Ng, A. Y. Rectifier nonlinearities improve neural network acoustic models. In Proc. 30th International Conference on Machine Learning (eds Dasgupta, S. & McAllester, D.) (JMLR, 2013).
Nair, V. & Hinton, G. E. Rectified linear units improve restricted Boltzmann machines. In Proc. 27th International Conference on Machine Learning 807–814 (Omnipress, 2010).
Ramachandran, P., Zoph, B. & Le, Q. V. Searching for activation functions. In Proc. 6th International Conference on Learning Representations (eds Murray, I., Ranzato, M. & Vinyals, O.) (ICLR, 2018).
Sutton, R. S. & Whitehead, S. D. Online learning with random representations. In Proc. 10th International Conference on Machine Learning 314–321 (Elsevier, 1993).
Lu, L., Shin, Y., Su, Y. & Karniadakis, G. E. Dying ReLU and initialization: theory and numerical examples. Commun. Computat. Phys . 28 , 1671–1706 (2020).
Shin, Y. & Karniadakis, G. E. Trainability of ReLU networks and data-dependent initialization. J. Mach. Learn. Model. Comput. 1 , 39–74 (2020).
Glorot, X. & Bengio, Y. Understanding the difficulty of training deep feedforward neural networks. In Proc. 13th International Conference on Artificial Intelligence and Statistics 249–256 (PMLR, 2010).
Montavon, G., Orr, G. & Müller, K.-R. Neural Networks: Tricks of the Trade (Springer, 2012).
Rakitianskaia, A. & Engelbrecht, A. Measuring saturation in neural networks. In Proc. 2015 IEEE Symposium Series on Computational Intelligence 1423–1430 (2015).
Boyd, S. P. & Vandenberghe, L. Convex Optimization (Cambridge Univ. Press, 2004).
Roy, O. & Vetterli, M. The effective rank: a measure of effective dimensionality. In Proc. 2007 15th European Signal Processing Conference 606–610 (IEEE, 2007).
Smith, S. L., Dherin, B., Barrett, D. & De, S. On the origin of implicit regularization in stochastic gradient descent. In Proc. 9th International Conference on Learning Representations (ICLR, 2021).
Razin, N. & Cohen, N. Implicit regularization in deep learning may not be explainable by norms. Adv. Neural Inf. Process. Syst. 33 , 21174–21187 (2020).
Goodfellow, I., Bengio, Y. & Courville, A. Deep Learning (MIT Press, 2016).
Srivastava, N., Hinton, G., Krizhevsky, A., Sutskever, I. & Salakhutdinov, R. Dropout: a simple way to prevent neural networks from overfitting. J. Mach. Learn. Res. 15 , 1929–1958 (2014).
Bjorck, N., Gomes, C. P., Selman, B. & Weinberger, K. Q. Understanding batch normalization. Adv. Neural Inf. Process. Syst. 31 , 7694–7705 (2018).
Ioffe, S. & Szegedy, C. Batch normalization: accelerating deep network training by reducing internal covariate shift. In Proc. 32nd International Conference on Machine Learning 448–456 (PMLR, 2015).
Chiley, V. et al. Online normalization for training neural networks. Adv. Neural Inf. Process. Syst. 32 , 1–11 (2019).
ADS Google Scholar
Ceron, J. S. O. & Castro, P. S. Revisiting rainbow: promoting more insightful and inclusive deep reinforcement learning research. In Proc. 38th International Conference on Machine Learning 1373–1383 (PMLR, 2021).
Brockman, G. et al. OpenAI Gym. Preprint at https://arxiv.org/abs/1606.01540 (2016).
Patterson, A., Neumann, S., White, M. & White, A. Empirical design in reinforcement learning. Preprint at https://arxiv.org/abs/2304.01315 (2023).
Igl, M., Farquhar, G., Luketina, J., Boehmer, W. & Whiteson, S. Transient non-stationarity and generalisation in deep reinforcement learning. In Proc. 9th International Conference on Learning Representations (ICLR, 2021).
Kumar, A., Agarwal, R., Ghosh, D. & Levine, S. Implicit under-parameterization inhibits data-efficient deep reinforcement learning. In Proc. 9th International Conference on Learning Representations (ICLR, 2021).
Nikishin, E. et al. Deep reinforcement learning with plasticity injection. Adv. Neural Inf. Process. Syst. 36 , 1–18 (2023).
D’Oro, P. et al. Sample-efficient reinforcement learning by breaking the replay ratio barrier. In Proc. 11th International Conference on Learning Representations (ICLR, 2023).
Schwarzer, M. et al. Bigger, better, faster: human-level Atari with human-level efficiency. In Proc. 40th International Conference on Machine Learning 30365–30380 (PMLR, 2023).
Lee, H. et al. PLASTIC: improving input and label plasticity for sample efficient reinforcement learning. Adv. Neural Inf. Process. Syst. 36 , 1–26 (2023).
CAS Google Scholar
Delfosse, Q., Schramowski, P., Mundt, M., Molina, A. & Kersting, K. Adaptive rational activations to boost deep reinforcement learning. In Proc. 12th International Conference on Learning Representations (ICLR, 2024).
Caruana, R. Multitask learning. Mach. Learn. 28 , 41–75 (1997).
Ring, M. B. in Learning to Learn (eds Thrun, S. & Pratt, L.) 261–292 (Springer, 1998).
Parisi, G. I., Kemker, R., Part, J. L., Kanan, C. & Wermter, S. Continual lifelong learning with neural networks: a review. Neural Netw. 113 , 54–71 (2019).
Kumar, S. et al. Continual learning as computationally constrained reinforcement learning. Preprint at https://arxiv.org/abs/2307.04345 (2023).
Yoon, J., Yang, E., Lee, J. & Hwang, S. J. Lifelong learning with dynamically expandable networks. In Proc. 6th International Conference on Learning Representations (eds Murray, I., Ranzato, M. & Vinyals, O.) (ICLR, 2018).
Aljundi, R. Online continual learning with maximal interfered retrieval. Adv. Neural Inf. Process. Syst. 32 , 1–12 (2019).
Golkar, S., Kagan, M. & Cho, K. in Proc. NeurIPS 2019 Workshop on Real Neurons & Hidden Units: Future Directions at the Intersection of Neuroscience and Artificial Intelligence 146 (NeurIPS, 2019).
Riemer, M. et al. Learning to learn without forgetting by maximizing transfer and minimizing interference. In Proc. 7th International Conference on Learning Representations (ICLR, 2019).
Rajasegaran, J., Hayat, M., Khan, S. H., Khan, F. & Shao, L. Random path selection for continual learning. Adv. Neural Inf. Process. Syst. 32 , 1–11 (2019).
Javed, K. & White, M. Meta-learning representations for continual learning. Adv. Neural Inf. Process. Syst. 32 , 1–11 (2019).
Veniat, T., Denoyer, L. & Ranzato, M. Efficient continual learning with modular networks and task-driven priors. In Proc. 9th International Conference on Learning Representations (ICLR, 2021).
Verwimp, E. et al. Continual learning: applications and the road forward. Trans. Mach. Learn. Res. https://openreview.net/forum?id=axBIMcGZn9 (2024).
Lopez-Paz, D. & Ranzato, M. Gradient episodic memory for continual learning. Adv. Neural Inf. Process. Syst. 30 , 1–10 (2017).
Rusu, A. A. et al. in Proc. 1st Annual Conference on Robot Learning 262–270 (PMLR, 2017).
Chen, J., Nguyen, T., Gorur, D. & Chaudhry, A. Is forgetting less a good inductive bias for forward transfer? In Proc. 11th International Conference on Learning Representations (ICLR, 2023).
Lewandowski, A., Tanaka, H., Schuurmans, D. & Machado, M. C. Directions of curvature as an explanation for loss of plasticity. Preprint at https://arxiv.org/abs/2312.00246 (2024).
Lyle, C. et al. Disentangling the causes of plasticity loss in neural networks. Preprint at https://arxiv.org/abs/2402.18762 (2024).
LeCun, Y., Denker, J. & Solla, S. Optimal brain damage. Adv. Neural Inf. Process. Syst. 2 , 598–605 (1989).
Han, S., Mao, H. & Dally, W. J. Deep compression: compressing deep neural networks with pruning, trained quantization and Huffman coding. In Proc. 4th International Conference on Learning Representations (ICLR, 2016).
Gale, T., Elsen, E. & Hooker, S. The state of sparsity in deep neural networks. Preprint at https://arxiv.org/abs/1902.09574 (2019).
Liu, J., Xu, Z., Shi, R., Cheung, R. C. C. & So, H. K. H. Dynamic sparse training: find efficient sparse network from scratch with trainable masked layers. In Proc. 8th International Conference on Learning Representations (ICLR, 2020).
Elsayed, M. & Mahmood, A. R. Addressing catastrophic forgetting and loss of plasticity in neural networks. In Proc. 12th International Conference on Learning Representations (ICLR, 2024).
Mocanu, D. C. et al. Scalable training of artificial neural networks with adaptive sparse connectivity inspired by network science. Nat. Commun. 9 , 2383 (2018).
Article ADS PubMed Google Scholar
Bellec, G., Kappel, D., Maass, W. & Legenstein, R. Deep rewiring: training very sparse deep networks. In Proc. 6th International Conference on Learning Representations (eds Murray, I., Ranzato, M. & Vinyals, O.) (ICLR, 2018).
Evci, U., Gale, T., Menick, J., Castro, P. S. & Elsen, E. Rigging the lottery: making all tickets winners. In Proc. 37th International Conference on Machine Learning 2943–2952 (PMLR, 2020).
Chen, T. et al. Chasing sparsity in vision transformers: an end-to-end exploration. Adv. Neural Inf. Process. Syst. 34 , 1–15 (2021).
Sokar, G., Mocanu, E., Mocanu, D. C., Pechenizkiy, M. & Stone, P. Dynamic sparse training for deep reinforcement learning. In Proc. 31st International Joint Conference on Artificial Intelligence (IJCAI-22) (ed. De Raedt, L.) 3437–3443 (IJCAI, 2022).
Graesser, L., Evci, U., Elsen, E. & Castro, P. S. The state of sparse training in deep reinforcement learning. In Proc. 39th International Conference on Machine Learning 7766–7792 (PMLR, 2022).
Zhou, G., Sohn, K. & Lee, H. Online incremental feature learning with denoising autoencoders. In Proc. 15th International Conference on Artificial Intelligence and Statistics 1453–1461 (PMLR, 2012).
Rusu, A. A. et al. Progressive neural networks. Preprint at https://arxiv.org/abs/1606.04671 (2022).
Sutskever, I., Martens, J., Dahl, G. & Hinton, G. On the importance of initialization and momentum in deep learning. In Proc. 30th International Conference on Machine Learning (eds Dasgupta, S. & McAllester, D.) 1139–1147 (JMLR, 2013).
Frankle, J. & Carbin, M. The lottery ticket hypothesis: finding sparse, trainable neural networks. In Proc. 7th International Conference on Learning Representations (ICLR, 2019).
Finn, C., Abbeel, P. & Levine, S. Model-agnostic meta-learning for fast adaptation of deep networks. In Proc. 34th International Conference on Machine Learning 1126–1135 (PMLR, 2017).
Wang, Y.-X., Ramanan, D. & Hebert, M. Growing a brain: fine-tuning by increasing model capacity. In Proc. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2471–2480 (IEEE, 2017).
Nagabandi, A. et al. Learning to adapt in dynamic, real-world environments through meta-reinforcement learning. In Proc. 7th International Conference on Learning Representations (ICLR, 2019).
Holmstrom, L. & Koistinen, P. et al. Using additive noise in back-propagation training. IEEE Trans. Neural Netw. 3 , 24–38 (1992).
Graves, A., Mohamed, A.-R. & Hinton, G. Speech recognition with deep recurrent neural networks. In Proc. 2013 IEEE International Conference on Acoustics, Speech and Signal Processing 6645–6649 (IEEE, 2013).
Neelakantan, A. et al. Adding gradient noise improves learning for very deep networks. Preprint at https://arxiv.org/abs/1511.06807 (2015).
Greff, K., Srivastava, R. K., Koutník, J., Steunebrink, B. R. & Schmidhuber, J. LSTM: a search space odyssey. IEEE Trans. Neural Netw. Learn. Syst. 28 , 2222–2232 (2017).
Article MathSciNet PubMed Google Scholar
Download references
Acknowledgements
We thank M. White for her feedback on an earlier version of this work; P. Nagarajan, E. Graves, G. Mihucz, A. Hakhverdyan, K. Roice, T. Ferguson, L. Watson, H. Sinha, P. Bhangale and M. Przystupa for their feedback on writing; and M. C. Machado for encouraging us to make this work accessible to a general scientific audience. We gratefully acknowledge the Digital Research Alliance of Canada for providing the computational resources to carry out the experiments in this paper. We also acknowledge funding from the Canada CIFAR AI Chairs program, DeepMind, the Alberta Machine Intelligence Institute (Amii), CIFAR and the Natural Sciences and Engineering Research Council of Canada (NSERC). This work was made possible by the stimulating and supportive research environment created by the members of the Reinforcement Learning and Artificial Intelligence (RLAI) laboratory, particularly within the agent-state research meetings.
Author information
Authors and affiliations.
Department of Computing Science, University of Alberta, Edmonton, Alberta, Canada
Shibhansh Dohare, J. Fernando Hernandez-Garcia, Qingfeng Lan, Parash Rahman, A. Rupam Mahmood & Richard S. Sutton
Canada CIFAR AI Chair, Alberta Machine Intelligence Institute (Amii), Edmonton, Alberta, Canada
A. Rupam Mahmood & Richard S. Sutton
You can also search for this author in PubMed Google Scholar
Contributions
S.D., J.F.H.-G., Q.L. and A.R.M. wrote the software. S.D., J.F.H.-G. and P.R. prepared the datasets. S.D. and J.F.H.-G. designed the experiments. S.D., J.F.H.-G., Q.L., R.S.S. and A.R.M. analysed and interpreted the results. S.D., A.R.M. and R.S.S. developed the continual backpropagation algorithm. S.D., J.F.H.-G., Q.L., R.S.S. and A.R.M. prepared the manuscript.
Corresponding author
Correspondence to Shibhansh Dohare .
Ethics declarations
Competing interests.
The authors declare no competing interests.
Peer review
Peer review information.
Nature thanks Pablo Castro, Razvan Pascanu and Gido van de Ven for their contribution to the peer review of this work. Peer reviewer reports are available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Extended data figures and tables
Extended data fig. 1 further results on class-incremental cifar-100..
a , Test accuracy in class-incremental CIFAR-100. As more classes are added, the classification becomes harder and algorithms naturally show decreasing accuracy with more classes. Each line corresponds to the average of 15 runs. b , Test accuracy of continual backpropagation for different values of the replacement rate parameter with contribution utility and 1,000 maturity threshold. The line corresponding to 10 −4 is an average of five runs, whereas the other two lines are an average of 15 runs. The solid lines represent the mean and the shaded regions correspond to ±1 standard error.
Extended Data Fig. 2 Loss of plasticity in the Slowly-Changing Regression problem.
a , The target function and the input in the Slowly-Changing Regression problem. The input has m + 1 bits. One of the flipping bits is chosen after every T time steps and its value is flipped. The next m − f bits are i.i.d. at every time step and the last bit is always one. The target function is represented by a neural network with a single hidden layer of LTUs. Each weight in the target network is −1 or 1. b , Loss of plasticity is robust across different activations. These results are averaged over 100 runs; the solid lines represent the mean and the shaded regions correspond to ±1 standard error.
Extended Data Fig. 3 Loss of plasticity in Online Permuted MNIST.
a , Left, an MNIST image with the label ‘7’; right, a corresponding permuted image. b , Loss of plasticity in Online Permuted MNIST is robust over step sizes, network sizes and rates of change. c , Evolution of various qualities of a deep network trained by means of backpropagation with different step sizes. Left, over time, the percentage of dead units in the network increases. Centre, the average magnitude of the weights increases over time. Right, the effective rank of the representation of the networks trained with backpropagation decreases over time. The results in these six plots are the average over 30 runs. The solid lines represent the mean and the shaded regions correspond to ±1 standard error. For some lines, the shaded region is thinner than the line width, as standard error is small.
Extended Data Fig. 4 Existing deep-learning methods on Online Permuted MNIST.
a , Left, online classification accuracy of various algorithms on Online Permuted MNIST. Shrink and Perturb has almost no drop in online classification accuracy over time. Continual backpropagation did not show any loss of plasticity and had the best level of performance. Centre left, over time, the percentage of dead units increases in all methods except for continual backpropagation; it has almost zero dead units throughout learning. Centre right, the average magnitude of the weights increases over time for all methods except for L2 regularization, Shrink and Perturb and continual backpropagation. These are also the three best-performing methods, which suggests that small weights are important for fast learning. Right, the effective rank of the representation of all methods drops over time. However, continual backpropagation maintains a higher effective rank than both backpropagation and Shrink and Perturb. Among all the algorithms, only continual backpropagation maintains a high effective rank, low weight magnitude and low percentage of dead units. The results correspond to the average over 30 independent runs. The shaded regions correspond to ±1 standard error. b , Performance of various algorithms on Online Permuted MNIST for various hyperparameter combinations. For each method, we show three different hyperparameter settings. The parameter settings that were used in the left panel in a are marked with a solid square next to their label. The results correspond to the average of over 30 runs for settings marked with a solid square and 10 runs for the rest. The solid lines represent the mean and the shaded regions correspond to ±1 standard error.
Extended Data Fig. 5 Further results in stationary reinforcement-learning problems.
a , Similar to Fig. 4 , the performance of standard PPO drops over time. However, unlike in Fig. 4 , the performance of PPO with L2 regularization gets worse over time in Hopper-v3. On the other hand, PPO with continual backpropagation and L2 regularization can keep improving with time. b , Comparison of continual backpropagation and ReDo on Ant-v3. The performance of PPO with ReDo and L2 regularization worsens over time, whereas PPO with continual backpropagation and L2 regularization keeps improving over time. c , PPO with standard Adam leads to large updates in the policy network compared with proper Adam ( β 1 = β 1 = 0.99), which explains why PPO with proper Adam performs much better than standard PPO. d , Comparison of two forms of utility in continual backpropagation, when using a running estimate of instantaneous utility and when using just the instantaneous utility. Both variations have similar performance. All these results are averaged over 30 runs; the solid lines represent the mean and the shaded regions correspond to 95% bootstrapped confidence interval.
Supplementary information
Peer review file, rights and permissions.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ .
Reprints and permissions
About this article
Cite this article.
Dohare, S., Hernandez-Garcia, J.F., Lan, Q. et al. Loss of plasticity in deep continual learning. Nature 632 , 768–774 (2024). https://doi.org/10.1038/s41586-024-07711-7
Download citation
Received : 11 August 2023
Accepted : 12 June 2024
Published : 21 August 2024
Issue Date : 22 August 2024
DOI : https://doi.org/10.1038/s41586-024-07711-7
Share this article
Anyone you share the following link with will be able to read this content:
Sorry, a shareable link is not currently available for this article.
Provided by the Springer Nature SharedIt content-sharing initiative
By submitting a comment you agree to abide by our Terms and Community Guidelines . If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.
Quick links
- Explore articles by subject
- Guide to authors
- Editorial policies
Sign up for the Nature Briefing: AI and Robotics newsletter — what matters in AI and robotics research, free to your inbox weekly.
![research paper about modular distance learning research paper about modular distance learning](https://verify.nature.com/verify/nature.png)
Information
- Author Services
Initiatives
You are accessing a machine-readable page. In order to be human-readable, please install an RSS reader.
All articles published by MDPI are made immediately available worldwide under an open access license. No special permission is required to reuse all or part of the article published by MDPI, including figures and tables. For articles published under an open access Creative Common CC BY license, any part of the article may be reused without permission provided that the original article is clearly cited. For more information, please refer to https://www.mdpi.com/openaccess .
Feature papers represent the most advanced research with significant potential for high impact in the field. A Feature Paper should be a substantial original Article that involves several techniques or approaches, provides an outlook for future research directions and describes possible research applications.
Feature papers are submitted upon individual invitation or recommendation by the scientific editors and must receive positive feedback from the reviewers.
Editor’s Choice articles are based on recommendations by the scientific editors of MDPI journals from around the world. Editors select a small number of articles recently published in the journal that they believe will be particularly interesting to readers, or important in the respective research area. The aim is to provide a snapshot of some of the most exciting work published in the various research areas of the journal.
Original Submission Date Received: .
- Active Journals
- Find a Journal
- Proceedings Series
- For Authors
- For Reviewers
- For Editors
- For Librarians
- For Publishers
- For Societies
- For Conference Organizers
- Open Access Policy
- Institutional Open Access Program
- Special Issues Guidelines
- Editorial Process
- Research and Publication Ethics
- Article Processing Charges
- Testimonials
- Preprints.org
- SciProfiles
- Encyclopedia
![Electronics electronics-logo](https://pub.mdpi-res.com/img/journals/electronics-logo.png?8600e93ff98dbf14)
Article Menu
![research paper about modular distance learning research paper about modular distance learning](https://www.mdpi.com/bundles/mdpisciprofileslink/img/unknown-user.png)
- Subscribe SciFeed
- Recommended Articles
- Google Scholar
- on Google Scholar
- Table of Contents
Find support for a specific problem in the support section of our website.
Please let us know what you think of our products and services.
Visit our dedicated information section to learn more about MDPI.
JSmol Viewer
A novel method for damping state switching based on machine learning of a strapdown inertial navigation system.
![ORCID research paper about modular distance learning](https://pub.mdpi-res.com/img/design/orcid.png?0465bc3812adeb52?1724941758)
1. Introduction
2. external horizontal damping network design, 2.1. principle of external horizontal damping network, 2.2. adaptive parameter design principle of damping network, 2.3. discrete method of damping network, 3. damping state identification method based on support vector machine, 3.1. the effect of reference velocity errors on sins, 3.2. principle of damping state switching method based on support vector machine detection, 3.3. overshoot error suppression principle, 4. system simulation test, 5. ship testing and verification, 6. conclusions, author contributions, data availability statement, conflicts of interest.
- Zhu, J.; Li, A.; Qin, F.; Chen, H.; Wang, J. A novel hybrid method based on deep learning for an integrated navigation system during DVL signal failure. Electronics 2022 , 11 , 2980. [ Google Scholar ] [ CrossRef ]
- Gao, D.; Hu, B.; Chang, L.; Qin, F.; Lyu, X. An Aided Navigation Method Based on Strapdown Gravity Gradiometer. Sensors 2021 , 21 , 829. [ Google Scholar ] [ CrossRef ]
- Mao, N.; Li, J.; He, H.; Xu, J. Attitude Estimation of SINS on Dynamic Base with Decoupling Gravity Vector. IEEE Sens. J. 2021 , 21 , 20530–20538. [ Google Scholar ] [ CrossRef ]
- Lyu, X.; Hu, B.; Li, K.; Chang, L. An Adaptive and Robust UKF Approach Based on Gaussian Process Regression-Aided Variational Bayesia. IEEE Sens. J. 2021 , 21 , 9500–9514. [ Google Scholar ] [ CrossRef ]
- Karmozdi, A.; Hashemi, M.; Salarieh, H.; Alasty, A. Implementation of Translational Motion Dynamics for INS Data Fusion in DVL Outage in Underwater Navigation. IEEE Sens. J. 2021 , 21 , 6652–6659. [ Google Scholar ] [ CrossRef ]
- Kudryashov, A.; Inzartsev, A. A Technique to Navigate Autonomous Underwater Vehicles Using a Virtual Coordinate Reference Network during Inspection of Industrial Subsea Structures. Remote Sens. 2022 , 14 , 5123. [ Google Scholar ] [ CrossRef ]
- Wang, D.; Xu, X.; Yao, Y.; Zhang, T. Virtual DVL Reconstruction Method for an Integrated Navigation System Based on DS-LSSVM Algorithm. IEEE Trans. Instrum. Meas. 2021 , 17 , 8501913. [ Google Scholar ] [ CrossRef ]
- Xu, B.; Wang, L.; Li, S.; Zhang, J. A Novel Calibration Method of SINS/DVL Integration Navigation System based on Quaternion. IEEE Sens. J. 2020 , 20 , 9567–9580. [ Google Scholar ] [ CrossRef ]
- Qin, Y. Inertial Navigation ; Beijing Science Press: Beijing, China, 2006. [ Google Scholar ]
- Liu, F.; Liu, C.; Weng, H.; Hu, X. Level damping algorithm of sins based on kalman filtering. J. Chin. Inert. Technol. 2013 , 21 , 285–288. [ Google Scholar ]
- Qin, F.; Li, A.; Xu, J. Improved internal damping method for inertial navigation system. J. China Inert. Technol. 2013 , 2 , 147–154. [ Google Scholar ]
- Li, Q.; Ben, Y.; Sun, F. A novel algorithm for marine strapdown gyrocompass based on digital filter. Measurement 2013 , 46 , 563–571. [ Google Scholar ] [ CrossRef ]
- Xu, B.; Wang, G.; Bai, J. Optimal design of damping network based on DVL velocity and IMU. Ocean Eng. 2017 , 132 , 101–113. [ Google Scholar ] [ CrossRef ]
- Kasper, J.F.; Nash, R.A. Doppler Radar Error Equations for Damped Inertial Navigation System Analysis. IEEE Trans. Aerosp. Electron. Syst. 1975 , 11 , 600–607. [ Google Scholar ] [ CrossRef ]
- Sun, J.; Xu, X.; Liu, Y.; Zhang, T.; Tong, J. An adaptive damping network designed for strapdown fiber optic gyrocompass system for ships. Sensors 2017 , 17 , 494. [ Google Scholar ] [ CrossRef ]
- Zhao, L.; Li, J.; Cheng, J.; Hao, Y. Damping strapdown inertial navigation system based on a kalman filter. Meas. Sci. Technol. 2016 , 27 , 115102. [ Google Scholar ] [ CrossRef ]
- Huang, W.; Fang, T.; Luo, L.; Zhao, L.; Che, F. A damping grid strapdown inertial navigation system based on a kalman filter for ships in polar regions. Sensors 2017 , 17 , 1551. [ Google Scholar ] [ CrossRef ]
- Qian, N.; Chang, G.; Guo, J.; Shen, W.; Yan, Z. Adaptive DDK Filter for GRACE Time-Variable Gravity Field with a Novel Anisotropic Filtering Strength Metric. Remote Sens. 2022 , 14 , 3114. [ Google Scholar ] [ CrossRef ]
- Zha, F.; Qin, F.; Li, F.; Ye, B. Fast external damping algorithm of inertial guidance system under external velocity reference condition. J. Wuhan Univ. 2019 , 44 , 398–404. [ Google Scholar ]
- Qin, F.; He, H.; Xu, J. Phase modulation-based sins damping method for autonomous vehicles. IEEE Sens. J. 2018 , 18 , 2483–2493. [ Google Scholar ]
- Qin, F.; Li, A.; Xu, J.; Zha, F. Horizontal inner damping method with continuously adjustable parameter for inertial navigation system. J. Chin. Inert. Technol. 2011 , 19 , 290–292+301. [ Google Scholar ]
- Yu, P.; Yang, G. An adaptive level damped algorithm based on PINS for ship. Adv. Mater. Res. 2013 , 760 , 2162–2166. [ Google Scholar ] [ CrossRef ]
- He, J.; Sun, C.; Zhang, B.; Wang, P. Variational Bayesian-based maximum correntropy cubature Kalman filter with both adaptivity and robustness. IEEE Sens. J. 2020 , 21 , 1982–1992. [ Google Scholar ] [ CrossRef ]
- Li, Q.; Ben, Y.; Sun, F. Strapdown fiber optic gyrocompass using adaptive network-based fuzzy inference system. Opt. Eng. 2014 , 53 , 014103. [ Google Scholar ] [ CrossRef ]
- Deng, Z.; Sun, J.; Ding, F.; Ismail, M.W.M. A novel damping method for strapdown inertial navigation system. IEEE Access 2019 , 7 , 49549–49557. [ Google Scholar ] [ CrossRef ]
- Liu, C.; Wu, Q.; Hu, P.; Zhang, R. Design of variable damping INS for ships based on the variation of reference velocity error. J. Sens. 2021 , 2021 , 6628073:1–6628073:15. [ Google Scholar ] [ CrossRef ]
- Qin, F.; Chen, Y.; Zha, F.; Chang, L. Marine Inertial Navigation Principle ; National Defence Industry Press: Beijing, China, 2018. [ Google Scholar ]
- Jiang, L.; Yu, Y.; Chen, Y. An adaptive-damping network designed for inertial navigation system of ships. Electron. Opt. Control 2014 , 21 , 52–57. [ Google Scholar ]
- Zhu, J.; Li, A.; Qin, F.; Chang, L. A New Robust Adaptive Filter Aided by Machine Learning Method for SINS/DVL Integrated Navigation System. Sensors 2022 , 22 , 3792. [ Google Scholar ] [ CrossRef ]
- Xu, J.; Zeng, W.; Lan, Y.; Guo, J.; Cheng, X. Modeling the parameter interactions in ranking svm with low-rank approximation. IEEE Trans. Knowl. Data Eng. 2019 , 31 , 1181–1193. [ Google Scholar ] [ CrossRef ]
- Wang, B.; Huang, L.; Liu, J.; Deng, Z.; Fu, M. A support vector regression-based integrated navigation method for underwater vehicles. IEEE Sens. J. 2020 , 20 , 8875–8883. [ Google Scholar ] [ CrossRef ]
- Xu, J.; He, H.; Qin, F.; Chang, L. A novel autonomous initial alignment method for strapdown inertial navigation system. IEEE Trans. Instrum. Meas. 2017 , 66 , 2274–2282. [ Google Scholar ] [ CrossRef ]
Click here to enlarge figure
Performance | Gyroscope | Accelerometer |
---|---|---|
Measuring Range | ±200 deg/s | ±15 g |
Update Rate | 200 Hz | 200 Hz |
Constant Drift | <0.02 deg/h | <50 μg |
Performance | DVL |
---|---|
Accuracy level | 0.5 %V ± 0.5 cm/s |
Speed range (backward movement is −) | −10 knot~20 knot |
Update rate | 1 Hz |
Operating rate | 300 Hz |
Bottom tracking depth | 300 m |
The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
Share and Cite
Lyu, X.; Zhu, J.; Wang, J.; Dong, R.; Qian, S.; Hu, B. A Novel Method for Damping State Switching Based on Machine Learning of a Strapdown Inertial Navigation System. Electronics 2024 , 13 , 3439. https://doi.org/10.3390/electronics13173439
Lyu X, Zhu J, Wang J, Dong R, Qian S, Hu B. A Novel Method for Damping State Switching Based on Machine Learning of a Strapdown Inertial Navigation System. Electronics . 2024; 13(17):3439. https://doi.org/10.3390/electronics13173439
Lyu, Xu, Jiupeng Zhu, Jungang Wang, Ruiqi Dong, Shiyi Qian, and Baiqing Hu. 2024. "A Novel Method for Damping State Switching Based on Machine Learning of a Strapdown Inertial Navigation System" Electronics 13, no. 17: 3439. https://doi.org/10.3390/electronics13173439
Article Metrics
Article access statistics, further information, mdpi initiatives, follow mdpi.
![MDPI Open Access Journals MDPI](https://pub.mdpi-res.com/img/design/mdpi-pub-logo-white-small.png?71d18e5f805839ab?1724941758)
Subscribe to receive issue release notifications and newsletters from MDPI journals
- Skip to main content
- Skip to search
- Skip to footer
Produkte und Services
![research paper about modular distance learning Catalyst 9300 Series Switches and Cisco Catalyst Center user interface on laptop screen](https://www.cisco.com/content/dam/cisco-cdc/site/images/heroes/products/networking/9300x-hero-overview-desktop-3200x1312.jpg)
Switches der Cisco Catalyst 9300-Serie
Ein sicherer arbeitsplatz – immer und überall.
Passen Sie Ihr Netzwerk an die Anforderungen der hybriden Arbeit an. Schaffen Sie eine sichere Grundlage, die optimierte Automatisierung und Einfachheit sowie umfassende Einblicke ermöglicht.
Machen Sie Ihr Netzwerk fit für die Zukunft
Ermöglichen Sie ein sichereres, nachhaltigeres Anwendungserlebnis, bei dem Geschwindigkeit und Skalierbarkeit mit erweiterter Automatisierung und detaillierten Analysen kombiniert werden.
Transformation für Ihren Arbeitsplatz
Erhalten Sie die Bandbreite, Geschwindigkeit und Skalierbarkeit, die Sie benötigen, um hybride Teams bei dem von ihnen gewählten Arbeitsmodell zu unterstützen.
Integrierte Netzwerksicherheit
Optimieren Sie Ihr Netzwerk mit integrierter Verschlüsselung und erweiterten KI-/ML-Lösungen für durchgängige Sicherheit nach dem Zero-Trust-Konzept.
Flexibles Netzwerkmanagement
Mit der Cisco Catalyst Center-Plattform oder dem Cloud-Monitoring für Catalyst können Sie herausragende Netzwerkerfahrungen bereitstellen.
Nachhaltiges Netzwerkdesign
Nutzen Sie das Netzwerk als Grundlage für intelligentere, nachhaltigere Gebäude, die energieeffizient sind und das Wohlbefinden der BewohnerInnen fördern.
Modelle der Catalyst 9300-Serie
C9300X-xx HX/HXN/TX SKUs
Catalyst 9300X (Kupfer)
Ausgelegt auf sicheren Hochgeschwindigkeits-Netzwerkzugriff, Aggregation und schlanke Zweigstellen mit 24 x oder 48 x 10-G-Multigigabit-Ports
- Optionen: nur Daten (kein PoE) und 90 W Cisco UPOE+
- Modulare Uplinks mit 100 G/40 G, 25 G/10 G/1 G oder 10 mGig
- Stacking-fähig bis zu 1 Tbit/s
C9300X-12Y/24Y SKUs
Catalyst 9300X (Glasfaser)
Ausgelegt auf sicheren Hochgeschwindigkeits-Netzwerkzugriff, Aggregation und Bereitstellungen für schlanke Zweigstellen
- 12 x oder 24 x Ports mit 25-G-/10-G-/1-G-Glasfaser
- Modulare Uplinks mit 100 G/40 G, 25 G/10 G/1 G oder 10 G mGig
C9300-24UB/UBX, C9300-48UB SKUs
Catalyst 9300 für besonders hohe Leistung
24 x oder 48 x Ports mit doppelten Puffern, MAC- und IP-Adressenskalierung und höherer ACL(Access Control List; Zugriffskontrolllisten)-Skalierung
- 1 G/2,5 G/5 G/10 G mGig und 90 W Cisco UPOE+
- Modulare Uplinks mit 1 G, 10 G, 25 G, 40 G oder 10 G mGig
- Stacking-fähig bis 480 Gbit/s
C9300-xx H/UX/UXM/UN SKUs
Catalyst 9300 UPOE+
Optimiert für konvergenten kabelgebundenen und Wireless-Zugriff, mit 24 oder 48 Ports
- 1 G oder 10 G Multigigabit mit 90 W Cisco UPOE+
C9300-xx T/P/U/S SKUs
Catalyst 9300 1 G
Grundlage für Intent-based Networking mit 24 x oder 48 x Ports mit 1 G für Daten
- PoE+, Cisco UPOE und SFP-Glasfaser
- Modulare Uplinks mit 1 G, 10 G, 25 G, 40 G oder Multigigabit
C9300L and C9300LM SKUs
Catalyst 9300L/LM 1 G
Ideal für den Einsatz in geschäftskritischen Zweigstellen mit wenig Platz
- 24 x oder 48 x Ports mit 1 G für Daten, PoE+ und Multigigabit
- Feste Uplinks mit 4 x 1 G, 4 x 10 G, 2 x 40 G
- Stacking-fähig bis 320 Gbit/s
![research paper about modular distance learning Cisco Catalyst Center-Schnittstelle](https://www.cisco.com/content/dam/cisco-cdc/site/us/en/images/networking/cat9300-switches-vod-thumbnail-1292x727.png)
Sehen Sie sich den Catalyst 9300X-Switch genauer an.
![research paper about modular distance learning Cisco Catalyst 9000 access point and Cisco Catalyst 9000 switch with Cisco Meraki dashboard](https://www.cisco.com/content/dam/cisco-cdc/site/images/graphics/products/networking/cloud-spotlight-promo-690x288.jpg)
Testen Sie unsere Cloud-Monitoring-Lösung für Ihr Switching-Netzwerk
So einfach und flexibel kann ortsunabhängiges Arbeiten sein – entdecken Sie unsere Cloud-Monitoring-Lösung für Cisco Catalyst Switches auf dem Cisco Meraki Dashboard.
Unterstützung während des gesamten Lebenszyklus
Netzwerkdienste
Beschleunigen Sie die Transformation Ihres Netzwerks
Erhalten Sie wertvolle Expertise und Einblicke für den Aufbau eines sicheren und flexiblen Netzwerks – vom Design bis hin zu Implementierung und Optimierung.
Supportservices
Ziehen Sie größeren Nutzen aus Ihren IT-Investitionen
Sorgen Sie mit fachkundiger Unterstützung dafür, dass Ihre IT und Ihr Unternehmen auf Kurs bleiben. Dazu bieten wir Ihnen rund um die Uhr Zugriff auf fundiertes technisches Wissen, umfassende globale Erfahrungen und digitale Intelligence, um Störungen und Risiken zu reduzieren.
Nehmen Sie an einer Demo teil
Vereinfachen Sie die Netzwerkverwaltung mit den Automatisierungs-, Assurance- und Analysefunktionen der Cisco Catalyst Center-Plattform. Nehmen Sie an einer Demo teil, um mehr darüber zu erfahren.
![](http://sektorel.online/777/templates/cheerup/res/banner1.gif)
IMAGES
VIDEO
COMMENTS
The term "modular approach" refers to learning that takes the form of individualized instruction and allows students to use Self-Learning Modules (SLMs) in the print or advanced format/electronic ...
The COVID-19 pandemic brought extraordinary challenges to K-12 students in using modular distance learning. According to Transactional Distance Theory (TDT), which is defined as understanding the effects of distance learning in the cognitive domain, the current study constructs a theoretical framework to measure student satisfaction and Bloom's Taxonomy Theory (BTT) to measure students ...
This mode of learning have been used by the learners and teachers during the conduct of Modular Distance Learning [3], [4]. This will attempt to increase their performances of the in the school. ...
ABSTRACT. Education in the new normal is a challenging task in the Philippines in an attempt to push through education amidst. the deadly pandemic caused by covid-19. The Department of Education ...
Academia.edu is a platform for academics to share research papers. Modular Distance Learning: a Phenomenological Study on Students' Challenges and Opportunities During Pandemic ... Document ID: PEMJ395, doi: 10.5281/zenodo.7394182, ISSN 2822-4353 Research Article Modular Distance Learning: a Phenomenological Study on Students' Challenges and ...
The purpose of this descriptive paper was to explore and synthesize literature related to understanding modular learning and how it can be implemented effectively so faculty members embrace its use. An in-depth review of literature addressed topics including, Educational Theories supporting modular learning, the development of modular learning ...
of modular distance learning or distance education resulted in a significant improvement ... (Cresswell 2014). This type of research allows the categorization of the results to date on the topic, as well as measuring the data based on different criteria regarding ... The search of the scientific papers was carried out in the international ...
100 The Implementation of Modular Distance Learning in the Philippine Secondary Public Schools 1Ysthr Rave Pe Dangle, 2Johnine Danganan Sumaoang 1Montessori Class Directress, British International School of Cambodia, Phnom Penh, Cambodia 2College Instructor, University of the Cordilleras, Baguio City, Philippines Abstract Face to face learning engagement of students and teachers within the ...
Domingo implemented Modular Distance Learning(MDL) in which a self-learning module(SLM) is the primary tool in the teaching and learning process. This study determines the effectiveness of the SLM in the implementation of MDL at the elementary level under the new normal. The study is anchored on Moore's Transactional Distance Theory.
207 Journal of Education, Teaching, and Learning Volume 6 Number 2 September 2021. Page 204-208 p-ISSN: 2477-5924 e-ISSN: 2477-8478 This research paper will help future researchers who will conduct future research about Modular Distance Learning (MDL).
We also found that the distance of their residences from the NEMSU campus slightly affected the percentage of modular learning material retrievals for both the BEED and BSED students, with R2 = 0.38 and R2 = 0.07, respectively. ... Feature papers represent the most advanced research with significant potential for high impact in the field ...
Among the distance learning modalities, this study focuses on modular distance learning as: 1) Applied in a broader coverage in the case country; 2) More inclusive with low- to ... Qualitative research is the most appropriate design of the study as it seeks to have a contextual understanding of phenomena, explain behavior and beliefs, identify ...
The existence of COVID-19 pandemic brought extraordinary challenges to the stakeholders, teachers, parents, and students. Thus, the researcher believed that there is an effect of teaching-learning process in new normal education to students' performance most especially using modular type of learning in Mathematics. With this, the study sought to determine the perception of the students ...
This paper viewed several articles discussing the role of modular distance learning modality in the performance of students, the positive and negative effects of MDL, and the parent's perception with regards on the implementation of the said modality ... Writing an abstract for a research paper | Manuscript formatting guidelines |
Experiences of Teachers on Using Modular Distance Learning (MDL) in Teaching Mathematics During the COVID-19 Pandemic. A. G. Roman. Education, Mathematics. - This study aims to present and analyze thematically the experiences of mathematics teachers in the implementation of modular distance learning in the first year of its implementation.
Abstract. This research aimed to gather data on the perceived effects of modular distance learning on students. The following research objectives informed this research: 1) to determine the perceived effects of modular distance learning on the students and 2) to determine the significant difference in the perceived effects of modular distance learning on grades 4,5 and 6 students.
Research Paper Open Access Teachers' Experiences in the Implementation of Modular Distance Learning Maricel C. Hidalgo Rofulo M. Landa High School, Philippines ABSTRACT:The continuity of learning in this time of pandemic brought challenges to many countries. This ... Modular Distance Learning is the most popular type of Distance Learning. It ...
Modular Distance Learning involves individualized instru ction that allows learners to use self- learning. modules (SLMs) in print or digital format/electronic copy, whichever is applicable in the ...
The study sought to identify the challenges faced by modular distance learning and the coping mechanisms. This was conducted on August 2022. Findings show that challenges include planning and preparing the modules, teaching the courses, collection of information on student performance, examined and assessed students' outputs, and provided feedback to students.
This study examined the impact of modular distance learning on students' motivation, interest/attitude, anxiety and achievement in mathematics. This was done at the Gabaldon, Nueva Ecija, Philippines during the first and second grading of the academic year 2021-2022. The study included both a descriptive-comparative and descriptive-correlational research design. The 207 high school students ...
This document provides an abstract and introduction for a research study on the effectiveness of modular distance learning for senior high school students in President Elpidio Quirino National High School during the 2020-2021 school year. Specifically, the study aims to determine the effectiveness, challenges, opinions, and recommendations of students regarding modular distance learning. The ...
This research study aims to identify the challenges and effectiveness of the modular distance learning approach on the academic performance of Grade 12 Humanities and Social Sciences students at ...
Modular Distance Learning offered a brand-new avenue in the education industry since the pandemic started. Amidst COVID-19 pandemic, DepEd ensured the continuity of learning through MDL-Self Learning Modules based on the Most Essential Learning Competencies. However, on the first few months of its implementation, several concerns where observed.
Machine learning and artificial intelligence have made remarkable progress in the past decade, with landmark successes in natural-language processing 2,3, biology 4, game playing 5,6,7,8 and ...
A Feature Paper should be a substantial original Article that involves several techniques or approaches, provides an outlook for future research directions and describes possible research applications. Feature papers are submitted upon individual invitation or recommendation by the scientific editors and must receive positive feedback from the ...
Switches der Cisco Catalyst 9300-Serie sind auf Sicherheit, IoT und die Cloud ausgelegt. Schaffen Sie eine sichere Grundlage, die optimierte Automatisierung und Einfachheit sowie umfassende Einblicke ermöglicht.