Adversarial Robustness for Unsupervised Domain Adaptation
Authors:
- 1 October 2021
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 8548-8557
Abstract
Extensive Unsupervised Domain Adaptation (UDA) studies have shown great success in practice by learning transferable representations across a labeled source domain and an unlabeled target domain with deep models. However, current work focuses on improving the generalization ability of UDA models on clean examples without considering the adversarial robustness, which is crucial in real-world applications. Conventional adversarial training methods are not suitable for the adversarial robustness on the unlabeled target domain of UDA since they train models with adversarial examples generated by the supervised loss function. In this work, we propose to leverage intermediate representations learned by robust ImageNet models to improve the robustness of UDA models. Our method works by aligning the features of the UDA model with the robust features learned by ImageNet pre-trained models along with domain adaptation training. It utilizes both labeled and unlabeled domains and instills robustness without any adversarial intervention or label requirement during domain adaptation training. Our experimental results show that our method significantly improves adversarial robustness compared to the baseline while keeping clean accuracy on various UDA benchmarks.Keywords
All Related Versions
This publication has 17 references indexed in Scilit:
- Domain Generalization by Solving Jigsaw Puzzles2019
- Do Better ImageNet Models Transfer Better?2019
- Defense Against Adversarial Attacks Using High-Level Representation Guided Denoiser2018
- Deeper, Broader and Artier Domain Generalization2017
- A Gift from Knowledge Distillation: Fast Optimization, Network Minimization and Transfer Learning2017
- Deep Hashing Network for Unsupervised Domain Adaptation2017
- Towards Evaluating the Robustness of Neural Networks2017
- Deep Residual Learning for Image Recognition2016
- Evasion Attacks against Machine Learning at Test Time2013
- A theory of learning from different domainsMachine Learning, 2009