Learning Bayesian Sparse Networks with Full Experience Replay for Continual Learning

Qingsen Yan, Dong Gong, Yuhang Liu, Anton Van Den Hengel, Javen Qinfeng Shi

科研成果: 书/报告/会议事项章节会议稿件同行评审

42 引用 (Scopus)

摘要

Continual Learning (CL) methods aim to enable machine learning models to learn new tasks without catastrophic forgetting of those that have been previously mastered. Existing CL approaches often keep a buffer of previously-seen samples, perform knowledge distillation, or use regularization techniques towards this goal. Despite their performance, they still suffer from interference across tasks which leads to catastrophic forgetting. To ameliorate this problem, we propose to only activate and select sparse neurons for learning current and past tasks at any stage. More parameters space and model capacity can thus be reserved for the future tasks. This minimizes the interference between parameters for different tasks. To do so, we propose a Sparse neural Network for Continual Learning (SNCL), which employs variational Bayesian sparsity priors on the activations of the neurons in all layers. Full Experience Replay (FER) provides effective supervision in learning the sparse activations of the neurons in different layers. A loss-aware reservoir-sampling strategy is developed to maintain the memory buffer. The proposed method is agnostic as to the network structures and the task boundaries. Experiments on different datasets show that SNCL achieves state-of-the-art result for mitigating forgetting.

源语言英语
主期刊名Proceedings - 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022
出版商IEEE Computer Society
109-118
页数10
ISBN(电子版)9781665469463
DOI
出版状态已出版 - 2022
已对外发布
活动2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022 - New Orleans, 美国
期限: 19 6月 202224 6月 2022

出版系列

姓名Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
2022-June
ISSN(印刷版)1063-6919

会议

会议2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022
国家/地区美国
New Orleans
时期19/06/2224/06/22

指纹

探究 'Learning Bayesian Sparse Networks with Full Experience Replay for Continual Learning' 的科研主题。它们共同构成独一无二的指纹。

引用此