Mitigating Feature Gap for Adversarial Robustness by Feature Disentanglement
Authors:
- Nuoyan ZhouNZNuoyan Zhou
- Dawei ZhouDZDawei Zhou
- Decheng LiuDLDecheng Liu
- Xinbo GaoXGXinbo Gao
- Nannan WangNWNannan Wang
Preprint
Abstract
Deep neural networks are vulnerable to adversarial samples. Adversarial fine-tuning methods aim to enhance adversarial robustness through fine-tuning the naturally pre-trained model in an adversarial training manner. However, we identify that some latent features of adversarial samples are confused by adversarial perturbation and lead to an unexpectedly increasing gap between features in the last hidden layer of natural and adversarial samples. To address this issue, we propose a disentanglement-based approach to explicitly model and further remove the latent features that cause the feature gap. Specifically, we introduce a feature disentangler to separate out the latent features from the features of the adversarial samples, thereby boosting robustness by eliminating the latent features. Besides, we align features in the pre-trained model with features of adversarial samples in the fine-tuned model, to further benefit from the features from natural samples without confusion. Empirical evaluations on three benchmark datasets demonstrate that our approach surpasses existing adversarial fine-tuning methods and adversarial training baselines.Keywords
All Related Versions
This publication has 7 references indexed in Scilit:
- Fast Adversarial Training With Adaptive Step SizeIEEE Transactions on Image Processing, 2023
- Adversarial Finetuning with Latent Representation Constraint to Mitigate Accuracy-Robustness Tradeoff2023
- Robust Evaluation of Diffusion-Based Adversarial Purification2023
- Robust Adversarial Attacks Detection Based on Explainable Deep Reinforcement Learning for UAV Guidance and PlanningIEEE Transactions on Intelligent Vehicles, 2023
- Better Robustness by More Coverage: Adversarial and Mixup Data Augmentation for Robust Finetuning2021
- Deep Residual Learning for Image Recognition2016
- Wide Residual Networks2016