Robust Generalization Against Photon-Limited Corruptions via Worst-Case Sharpness Minimization

Zhuo Huang, Miaoxi Zhu, Xiaobo Xia, Li Shen, Jun Yu*, Chen Gong, Bo Han, Bo Du, Tongliang Liu

*Corresponding author for this work

Research output: Chapter in book/report/conference proceedingConference proceedingpeer-review

1 Citation (Scopus)

Abstract

Robust generalization aims to tackle the most challenging data distributions which are rare in the training set and contain severe noises, i.e., photon-limited corruptions. Common solutions such as distributionally robust optimization (DRO) focus on the worst-case empirical risk to ensure low training error on the uncommon noisy distributions. However, due to the over-parameterized model being optimized on scarce worst-case data, DRO fails to produce a smooth loss landscape, thus struggling on generalizing well to the test set. Therefore, instead of focusing on the worst-case risk minimization, we propose SharpDRO by penalizing the sharpness of the worst-case distribution, which measures the loss changes around the neighbor of learning parameters. Through worst-case sharpness minimization, the proposed method successfully produces a flat loss curve on the corrupted distributions, thus achieving robust generalization. Moreover, by considering whether the distribution annotation is available, we apply SharpDRO to two problem settings and design a worst-case selection process for robust generalization. Theoretically, we show that SharpDRO has a great convergence guarantee. Experimentally, we simulate photon-limited corruptions using CIFAR10/100 and ImageNet30 datasets and show that SharpDRO exhibits a strong generalization ability against severe corruptions and exceeds well-known baseline methods with large performance gains.

Original languageEnglish
Title of host publicationProceedings - 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2023
Place of PublicationVancouver, BC, Canada
PublisherIEEE
Pages16175-16185
Number of pages11
ISBN (Electronic)9798350301298
ISBN (Print)9798350301304
DOIs
Publication statusPublished - 17 Jun 2023
Event36th IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2023 - Vancouver, Canada
Duration: 17 Jun 202322 Jun 2023
https://cvpr2023.thecvf.com/
https://openaccess.thecvf.com/CVPR2023
https://ieeexplore.ieee.org/xpl/conhome/10203037/proceeding

Publication series

NameProceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
Volume2023-June
ISSN (Print)1063-6919
ISSN (Electronic)2575-7075

Conference

Conference36th IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2023
Country/TerritoryCanada
CityVancouver
Period17/06/2322/06/23
Internet address

Scopus Subject Areas

  • Software
  • Computer Vision and Pattern Recognition

User-Defined Keywords

  • Optimization methods (other than deep learning)

Fingerprint

Dive into the research topics of 'Robust Generalization Against Photon-Limited Corruptions via Worst-Case Sharpness Minimization'. Together they form a unique fingerprint.

Cite this