Timezone: »
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.
Author Information
Zhenyi Wang (University at Buffalo)
Li Shen (JD Explore Academy)
Le Fang (University at Buffalo-SUNY)
Qiuling Suo (State University of New York at Buffalo)
Tiehang Duan (State University of New York at Buffalo)
Mingchen Gao (University at Buffalo, SUNY)
Related Events (a corresponding poster, oral, or spotlight)
-
2022 Poster: Improving Task-free Continual Learning by Distributionally Robust Memory Evolution »
Tue. Jul 19th through Wed the 20th Room Hall E #518
More from the Same Authors
-
2023 Poster: Dynamic Regularized Sharpness Aware Minimization in Federated Learning: Approaching Global Consistency and Smooth Landscape »
Yan Sun · Li Shen · Shixiang Chen · Liang Ding · Dacheng Tao -
2023 Poster: Are Large Kernels Better Teachers than Transformers for ConvNets? »
Tianjin Huang · Lu Yin · Zhenyu Zhang · Li Shen · Meng Fang · Mykola Pechenizkiy · Zhangyang “Atlas” Wang · Shiwei Liu -
2023 Poster: Learning to Learn from APIs: Black-box Data-free Meta-Learning »
Zixuan Hu · Li Shen · Zhenyi Wang · Baoyuan Wu · Chun Yuan · Dacheng Tao -
2023 Poster: Improving the Model Consistency of Decentralized Federated Learning »
Yifan Shi · Li Shen · Kang Wei · Yan Sun · Bo Yuan · Xueqian Wang · Dacheng Tao -
2023 Poster: CoCo: A Coupled Contrastive Framework for Unsupervised Domain Adaptive Graph Classification »
Nan Yin · Li Shen · Mengzhu Wang · Long Lan · Zeyu Ma · Chong Chen · Xian-Sheng Hua · Xiao Luo -
2023 Oral: Dynamic Regularized Sharpness Aware Minimization in Federated Learning: Approaching Global Consistency and Smooth Landscape »
Yan Sun · Li Shen · Shixiang Chen · Liang Ding · Dacheng Tao -
2022 : Paper 12: SafeRL-Kit: Evaluating Efficient Reinforcement Learning Methods for Safe Autonomous Driving »
· Li Shen · Bo Yuan · Xueqian Wang -
2022 Poster: Understanding Robust Overfitting of Adversarial Training and Beyond »
Chaojian Yu · Bo Han · Li Shen · Jun Yu · Chen Gong · Mingming Gong · Tongliang Liu -
2022 Poster: DisPFL: Towards Communication-Efficient Personalized Federated Learning via Decentralized Sparse Training »
Rong Dai · Li Shen · Fengxiang He · Xinmei Tian · Dacheng Tao -
2022 Spotlight: Understanding Robust Overfitting of Adversarial Training and Beyond »
Chaojian Yu · Bo Han · Li Shen · Jun Yu · Chen Gong · Mingming Gong · Tongliang Liu -
2022 Spotlight: DisPFL: Towards Communication-Efficient Personalized Federated Learning via Decentralized Sparse Training »
Rong Dai · Li Shen · Fengxiang He · Xinmei Tian · Dacheng Tao -
2022 Poster: Deep Neural Network Fusion via Graph Matching with Applications to Model Ensemble and Federated Learning »
Chang Liu · Chenfei Lou · Runzhong Wang · Alan Yuhan Xi · Li Shen · Junchi Yan -
2022 Spotlight: Deep Neural Network Fusion via Graph Matching with Applications to Model Ensemble and Federated Learning »
Chang Liu · Chenfei Lou · Runzhong Wang · Alan Yuhan Xi · Li Shen · Junchi Yan -
2020 Poster: Communication-Efficient Distributed Stochastic AUC Maximization with Deep Neural Networks »
Zhishuai Guo · Mingrui Liu · Zhuoning Yuan · Li Shen · Wei Liu · Tianbao Yang -
2018 Poster: An Algorithmic Framework of Variable Metric Over-Relaxed Hybrid Proximal Extra-Gradient Method »
Li Shen · Peng Sun · Yitong Wang · Wei Liu · Tong Zhang -
2018 Oral: An Algorithmic Framework of Variable Metric Over-Relaxed Hybrid Proximal Extra-Gradient Method »
Li Shen · Peng Sun · Yitong Wang · Wei Liu · Tong Zhang -
2017 Poster: GSOS: Gauss-Seidel Operator Splitting Algorithm for Multi-Term Nonsmooth Convex Composite Optimization »
Li Shen · Wei Liu · Ganzhao Yuan · Shiqian Ma -
2017 Talk: GSOS: Gauss-Seidel Operator Splitting Algorithm for Multi-Term Nonsmooth Convex Composite Optimization »
Li Shen · Wei Liu · Ganzhao Yuan · Shiqian Ma