Consistency is the Key to Further Mitigating Catastrophic Forgetting in Continual Learning

Prashant Shivaram Bhat, Bahram Zonooz, Elahe Arani
Proceedings of The 1st Conference on Lifelong Learning Agents, PMLR 199:1195-1212, 2022.

Abstract

Deep neural networks struggle to continually learn multiple sequential tasks due to catastrophic forgetting of previously learned tasks. Rehearsal-based methods which explicitly store previous task samples in the buffer and interleave them with the current task samples have proven to be the most effective in mitigating forgetting. However, Experience Replay (ER) does not perform well under low-buffer regimes and longer task sequences as its performance is commensurate with the buffer size. Consistency in predictions of soft-targets can assist ER in preserving information pertaining to previous tasks better as soft-targets capture the rich similarity structure of the data. Therefore, we examine the role of consistency regularization in ER framework under various continual learning scenarios. We also propose to cast consistency regularization as a self-supervised pretext task thereby enabling the use of a wide variety of self-supervised learning methods as regularizers. While simultaneously enhancing model calibration and robustness to natural corruptions, regularizing consistency in predictions results in lesser forgetting across all continual learning scenarios. Among the different families of regularizers, we find that stricter consistency constraints preserve previous task information in ER better.

Cite this Paper


BibTeX
@InProceedings{pmlr-v199-bhat22b, title = {Consistency is the Key to Further Mitigating Catastrophic Forgetting in Continual Learning}, author = {Bhat, Prashant Shivaram and Zonooz, Bahram and Arani, Elahe}, booktitle = {Proceedings of The 1st Conference on Lifelong Learning Agents}, pages = {1195--1212}, year = {2022}, editor = {Chandar, Sarath and Pascanu, Razvan and Precup, Doina}, volume = {199}, series = {Proceedings of Machine Learning Research}, month = {22--24 Aug}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v199/bhat22b/bhat22b.pdf}, url = {https://proceedings.mlr.press/v199/bhat22b.html}, abstract = {Deep neural networks struggle to continually learn multiple sequential tasks due to catastrophic forgetting of previously learned tasks. Rehearsal-based methods which explicitly store previous task samples in the buffer and interleave them with the current task samples have proven to be the most effective in mitigating forgetting. However, Experience Replay (ER) does not perform well under low-buffer regimes and longer task sequences as its performance is commensurate with the buffer size. Consistency in predictions of soft-targets can assist ER in preserving information pertaining to previous tasks better as soft-targets capture the rich similarity structure of the data. Therefore, we examine the role of consistency regularization in ER framework under various continual learning scenarios. We also propose to cast consistency regularization as a self-supervised pretext task thereby enabling the use of a wide variety of self-supervised learning methods as regularizers. While simultaneously enhancing model calibration and robustness to natural corruptions, regularizing consistency in predictions results in lesser forgetting across all continual learning scenarios. Among the different families of regularizers, we find that stricter consistency constraints preserve previous task information in ER better.} }
Endnote
%0 Conference Paper %T Consistency is the Key to Further Mitigating Catastrophic Forgetting in Continual Learning %A Prashant Shivaram Bhat %A Bahram Zonooz %A Elahe Arani %B Proceedings of The 1st Conference on Lifelong Learning Agents %C Proceedings of Machine Learning Research %D 2022 %E Sarath Chandar %E Razvan Pascanu %E Doina Precup %F pmlr-v199-bhat22b %I PMLR %P 1195--1212 %U https://proceedings.mlr.press/v199/bhat22b.html %V 199 %X Deep neural networks struggle to continually learn multiple sequential tasks due to catastrophic forgetting of previously learned tasks. Rehearsal-based methods which explicitly store previous task samples in the buffer and interleave them with the current task samples have proven to be the most effective in mitigating forgetting. However, Experience Replay (ER) does not perform well under low-buffer regimes and longer task sequences as its performance is commensurate with the buffer size. Consistency in predictions of soft-targets can assist ER in preserving information pertaining to previous tasks better as soft-targets capture the rich similarity structure of the data. Therefore, we examine the role of consistency regularization in ER framework under various continual learning scenarios. We also propose to cast consistency regularization as a self-supervised pretext task thereby enabling the use of a wide variety of self-supervised learning methods as regularizers. While simultaneously enhancing model calibration and robustness to natural corruptions, regularizing consistency in predictions results in lesser forgetting across all continual learning scenarios. Among the different families of regularizers, we find that stricter consistency constraints preserve previous task information in ER better.
APA
Bhat, P.S., Zonooz, B. & Arani, E.. (2022). Consistency is the Key to Further Mitigating Catastrophic Forgetting in Continual Learning. Proceedings of The 1st Conference on Lifelong Learning Agents, in Proceedings of Machine Learning Research 199:1195-1212 Available from https://proceedings.mlr.press/v199/bhat22b.html.

Related Material