Timezone: »
Test-time adaptation provides an effective means of tackling the potential distribution shift between model training and inference, by dynamically updating the model at test time. This area has seen fast progress recently, at the effectiveness of handling test shifts. Nonetheless, prior methods still suffer two key limitations: 1) these methods rely on performing backward computation for each test sample, which takes a considerable amount of time; and 2) these methods focus on improving the performance on out-of-distribution test samples and ignore that the adaptation on test data may result in a catastrophic forgetting issue, \ie, the performance on in-distribution test samples may degrade. To address these issues, we propose an efficient anti-forgetting test-time adaptation (EATA) method. Specifically, we devise a sample-efficient entropy minimization loss to exclude uninformative samples out of backward computation, which improves the overall efficiency and meanwhile boosts the out-of-distribution accuracy. Afterward, we introduce a regularization loss to ensure that critical model weights tend to be preserved during adaptation, thereby alleviating the forgetting issue. Extensive experiments on CIFAR-10-C, ImageNet-C, and ImageNet-R verify the effectiveness and superiority of our EATA.
Author Information
Shuaicheng Niu (South China University of Technology)
Jiaxiang Wu (Tencent AI Lab)
Yifan Zhang (National University of Singapore)
Yaofo Chen (South China University of Technology)
Shijian Zheng (South China University of Technology)
Peilin Zhao (Tencent AI Lab)
Mingkui Tan (South China University of Technology)
Related Events (a corresponding poster, oral, or spotlight)
-
2022 Poster: Efficient Test-Time Model Adaptation without Forgetting »
Thu. Jul 21st through Fri the 22nd Room Hall E #529
More from the Same Authors
-
2023 Poster: Detecting Adversarial Data by Probing Multiple Perturbations Using Expected Perturbation Score »
Shuhai Zhang · Feng Liu · Jiahao Yang · 逸凡 杨 · Changsheng Li · Bo Han · Mingkui Tan -
2022 Poster: $p$-Laplacian Based Graph Neural Networks »
Guoji Fu · Peilin Zhao · Yatao Bian -
2022 Poster: Local Augmentation for Graph Neural Networks »
Songtao Liu · Rex (Zhitao) Ying · Hanze Dong · Lanqing Li · Tingyang Xu · Yu Rong · Peilin Zhao · Junzhou Huang · Dinghao Wu -
2022 Spotlight: $p$-Laplacian Based Graph Neural Networks »
Guoji Fu · Peilin Zhao · Yatao Bian -
2022 Spotlight: Local Augmentation for Graph Neural Networks »
Songtao Liu · Rex (Zhitao) Ying · Hanze Dong · Lanqing Li · Tingyang Xu · Yu Rong · Peilin Zhao · Junzhou Huang · Dinghao Wu -
2021 Poster: AdaXpert: Adapting Neural Architecture for Growing Data »
Shuaicheng Niu · Jiaxiang Wu · Guanghui Xu · Yifan Zhang · Yong Guo · Peilin Zhao · Peng Wang · Mingkui Tan -
2021 Spotlight: AdaXpert: Adapting Neural Architecture for Growing Data »
Shuaicheng Niu · Jiaxiang Wu · Guanghui Xu · Yifan Zhang · Yong Guo · Peilin Zhao · Peng Wang · Mingkui Tan -
2021 Poster: Meta-learning Hyperparameter Performance Prediction with Neural Processes »
Ying WEI · Peilin Zhao · Junzhou Huang -
2021 Spotlight: Meta-learning Hyperparameter Performance Prediction with Neural Processes »
Ying WEI · Peilin Zhao · Junzhou Huang -
2020 Poster: Breaking the Curse of Space Explosion: Towards Efficient NAS with Curriculum Search »
Yong Guo · Yaofo Chen · Yin Zheng · Peilin Zhao · Jian Chen · Junzhou Huang · Mingkui Tan -
2019 Poster: Collaborative Channel Pruning for Deep Networks »
Hanyu Peng · Jiaxiang Wu · Shifeng Chen · Junzhou Huang -
2019 Oral: Collaborative Channel Pruning for Deep Networks »
Hanyu Peng · Jiaxiang Wu · Shifeng Chen · Junzhou Huang -
2018 Poster: Adversarial Learning with Local Coordinate Coding »
Jiezhang Cao · Yong Guo · Qingyao Wu · Chunhua Shen · Junzhou Huang · Mingkui Tan -
2018 Poster: Error Compensated Quantized SGD and its Applications to Large-scale Distributed Optimization »
Jiaxiang Wu · Weidong Huang · Junzhou Huang · Tong Zhang -
2018 Oral: Adversarial Learning with Local Coordinate Coding »
Jiezhang Cao · Yong Guo · Qingyao Wu · Chunhua Shen · Junzhou Huang · Mingkui Tan -
2018 Oral: Error Compensated Quantized SGD and its Applications to Large-scale Distributed Optimization »
Jiaxiang Wu · Weidong Huang · Junzhou Huang · Tong Zhang