Balance, Imbalance, and Rebalance: Understanding Robust Overfitting from a Minimax Game Perspective

• URL: http://arxiv.org/abs/2310.19360v1
• Date: Mon, 30 Oct 2023 09:00:11 GMT
• Title: Balance, Imbalance, and Rebalance: Understanding Robust Overfitting from a Minimax Game Perspective
• Authors: Yifei Wang, Liangchen Li, Jiansheng Yang, Zhouchen Lin, Yisen Wang
• Abstract summary: Adversarial Training (AT) has become arguably the state-of-the-art algorithm for extracting robust features. AT suffers from severe robust overfitting problems, particularly after learning rate (LR) decay. We show how LR decay breaks the balance between the minimax game by empowering the trainer with a stronger memorization ability.
• Score: 80.51463286812314
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Adversarial Training (AT) has become arguably the state-of-the-art algorithm for extracting robust features. However, researchers recently notice that AT suffers from severe robust overfitting problems, particularly after learning rate (LR) decay. In this paper, we explain this phenomenon by viewing adversarial training as a dynamic minimax game between the model trainer and the attacker. Specifically, we analyze how LR decay breaks the balance between the minimax game by empowering the trainer with a stronger memorization ability, and show such imbalance induces robust overfitting as a result of memorizing non-robust features. We validate this understanding with extensive experiments, and provide a holistic view of robust overfitting from the dynamics of both the two game players. This understanding further inspires us to alleviate robust overfitting by rebalancing the two players by either regularizing the trainer's capacity or improving the attack strength. Experiments show that the proposed ReBalanced Adversarial Training (ReBAT) can attain good robustness and does not suffer from robust overfitting even after very long training. Code is available at https://github.com/PKU-ML/ReBAT.

Related papers

• Randomized Adversarial Training via Taylor Expansion [18.54106339075049]
  We propose a novel adversarial training method via Taylor expansion of a small Gaussian noise. We show that the new adversarial training method can flatten loss landscape and find flat minima. With PGD, CW, and Auto Attacks, an extensive set of experiments demonstrate that our method boosts both robustness and clean accuracy.
  arXiv  Detail & Related papers  (2023-03-19T13:30:33Z)
• Nash Equilibria and Pitfalls of Adversarial Training in Adversarial Robustness Games [51.90475640044073]
  We study adversarial training as an alternating best-response strategy in a 2-player zero-sum game. On the other hand, a unique pure Nash equilibrium of the game exists and is provably robust.
  arXiv  Detail & Related papers  (2022-10-23T03:21:01Z)
• Enhancing Adversarial Training with Feature Separability [52.39305978984573]
  We introduce a new concept of adversarial training graph (ATG) with which the proposed adversarial training with feature separability (ATFS) enables to boost the intra-class feature similarity and increase inter-class feature variance. Through comprehensive experiments, we demonstrate that the proposed ATFS framework significantly improves both clean and robust performance.
  arXiv  Detail & Related papers  (2022-05-02T04:04:23Z)
• Imbalanced Adversarial Training with Reweighting [33.51820466479575]
  We show that adversarially trained models can suffer much worse performance on under-represented classes, when the training dataset is imbalanced. Traditional reweighting strategies may lose efficacy to deal with the imbalance issue for adversarial training. We propose Separable Reweighted Adversarial Training (SRAT) to facilitate adversarial training under imbalanced scenarios.
  arXiv  Detail & Related papers  (2021-07-28T20:51:36Z)
• Robust Pre-Training by Adversarial Contrastive Learning [120.33706897927391]
  Recent work has shown that, when integrated with adversarial training, self-supervised pre-training can lead to state-of-the-art robustness. We improve robustness-aware self-supervised pre-training by learning representations consistent under both data augmentations and adversarial perturbations.
  arXiv  Detail & Related papers  (2020-10-26T04:44:43Z)
• Towards Understanding Fast Adversarial Training [91.8060431517248]
  We conduct experiments to understand the behavior of fast adversarial training. We show the key to its success is the ability to recover from overfitting to weak attacks.
  arXiv  Detail & Related papers  (2020-06-04T18:19:43Z)
• Overfitting in adversarially robust deep learning [86.11788847990783]
  We show that overfitting to the training set does in fact harm robust performance to a very large degree in adversarially robust training. We also show that effects such as the double descent curve do still occur in adversarially trained models, yet fail to explain the observed overfitting.
  arXiv  Detail & Related papers  (2020-02-26T15:40:50Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.