Improving Task-free Continual Learning by Distributionally Robust Memory Evolution

Zhenyi Wang, Li Shen, Le Fang, Qiuling Suo, Tiehang Duan, Mingchen Gao
Proceedings of the 39th International Conference on Machine Learning, PMLR 162:22985-22998, 2022.

Abstract

Task-free continual learning (CL) aims to learn a non-stationary data stream without explicit task definitions and not forget previous knowledge. The widely adopted memory replay approach could gradually become less effective for long data streams, as the model may memorize the stored examples and overfit the memory buffer. Second, existing methods overlook the high uncertainty in the memory data distribution since there is a big gap between the memory data distribution and the distribution of all the previous data examples. To address these problems, for the first time, we propose a principled memory evolution framework to dynamically evolve the memory data distribution by making the memory buffer gradually harder to be memorized with distributionally robust optimization (DRO). We then derive a family of methods to evolve the memory buffer data in the continuous probability measure space with Wasserstein gradient flow (WGF). The proposed DRO is w.r.t the worst-case evolved memory data distribution, thus guarantees the model performance and learns significantly more robust features than existing memory-replay-based methods. Extensive experiments on existing benchmarks demonstrate the effectiveness of the proposed methods for alleviating forgetting. As a by-product of the proposed framework, our method is more robust to adversarial examples than existing task-free CL methods.

Cite this Paper


BibTeX
@InProceedings{pmlr-v162-wang22v, title = {Improving Task-free Continual Learning by Distributionally Robust Memory Evolution}, author = {Wang, Zhenyi and Shen, Li and Fang, Le and Suo, Qiuling and Duan, Tiehang and Gao, Mingchen}, booktitle = {Proceedings of the 39th International Conference on Machine Learning}, pages = {22985--22998}, year = {2022}, editor = {Chaudhuri, Kamalika and Jegelka, Stefanie and Song, Le and Szepesvari, Csaba and Niu, Gang and Sabato, Sivan}, volume = {162}, series = {Proceedings of Machine Learning Research}, month = {17--23 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v162/wang22v/wang22v.pdf}, url = {https://proceedings.mlr.press/v162/wang22v.html}, abstract = {Task-free continual learning (CL) aims to learn a non-stationary data stream without explicit task definitions and not forget previous knowledge. The widely adopted memory replay approach could gradually become less effective for long data streams, as the model may memorize the stored examples and overfit the memory buffer. Second, existing methods overlook the high uncertainty in the memory data distribution since there is a big gap between the memory data distribution and the distribution of all the previous data examples. To address these problems, for the first time, we propose a principled memory evolution framework to dynamically evolve the memory data distribution by making the memory buffer gradually harder to be memorized with distributionally robust optimization (DRO). We then derive a family of methods to evolve the memory buffer data in the continuous probability measure space with Wasserstein gradient flow (WGF). The proposed DRO is w.r.t the worst-case evolved memory data distribution, thus guarantees the model performance and learns significantly more robust features than existing memory-replay-based methods. Extensive experiments on existing benchmarks demonstrate the effectiveness of the proposed methods for alleviating forgetting. As a by-product of the proposed framework, our method is more robust to adversarial examples than existing task-free CL methods.} }
Endnote
%0 Conference Paper %T Improving Task-free Continual Learning by Distributionally Robust Memory Evolution %A Zhenyi Wang %A Li Shen %A Le Fang %A Qiuling Suo %A Tiehang Duan %A Mingchen Gao %B Proceedings of the 39th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2022 %E Kamalika Chaudhuri %E Stefanie Jegelka %E Le Song %E Csaba Szepesvari %E Gang Niu %E Sivan Sabato %F pmlr-v162-wang22v %I PMLR %P 22985--22998 %U https://proceedings.mlr.press/v162/wang22v.html %V 162 %X Task-free continual learning (CL) aims to learn a non-stationary data stream without explicit task definitions and not forget previous knowledge. The widely adopted memory replay approach could gradually become less effective for long data streams, as the model may memorize the stored examples and overfit the memory buffer. Second, existing methods overlook the high uncertainty in the memory data distribution since there is a big gap between the memory data distribution and the distribution of all the previous data examples. To address these problems, for the first time, we propose a principled memory evolution framework to dynamically evolve the memory data distribution by making the memory buffer gradually harder to be memorized with distributionally robust optimization (DRO). We then derive a family of methods to evolve the memory buffer data in the continuous probability measure space with Wasserstein gradient flow (WGF). The proposed DRO is w.r.t the worst-case evolved memory data distribution, thus guarantees the model performance and learns significantly more robust features than existing memory-replay-based methods. Extensive experiments on existing benchmarks demonstrate the effectiveness of the proposed methods for alleviating forgetting. As a by-product of the proposed framework, our method is more robust to adversarial examples than existing task-free CL methods.
APA
Wang, Z., Shen, L., Fang, L., Suo, Q., Duan, T. & Gao, M.. (2022). Improving Task-free Continual Learning by Distributionally Robust Memory Evolution. Proceedings of the 39th International Conference on Machine Learning, in Proceedings of Machine Learning Research 162:22985-22998 Available from https://proceedings.mlr.press/v162/wang22v.html.

Related Material