LAS-AT: Adversarial Training with Learnable Attack Strategy

by   Xiaojun Jia, et al.
University of the Chinese Academy of Sciences
The Chinese University of Hong Kong, Shenzhen

Adversarial training (AT) is always formulated as a minimax problem, of which the performance depends on the inner optimization that involves the generation of adversarial examples (AEs). Most previous methods adopt Projected Gradient Decent (PGD) with manually specifying attack parameters for AE generation. A combination of the attack parameters can be referred to as an attack strategy. Several works have revealed that using a fixed attack strategy to generate AEs during the whole training phase limits the model robustness and propose to exploit different attack strategies at different training stages to improve robustness. But those multi-stage hand-crafted attack strategies need much domain expertise, and the robustness improvement is limited. In this paper, we propose a novel framework for adversarial training by introducing the concept of "learnable attack strategy", dubbed LAS-AT, which learns to automatically produce attack strategies to improve the model robustness. Our framework is composed of a target network that uses AEs for training to improve robustness and a strategy network that produces attack strategies to control the AE generation. Experimental evaluations on three benchmark databases demonstrate the superiority of the proposed method. The code is released at


page 1

page 2

page 3

page 4


CAT:Collaborative Adversarial Training

Adversarial training can improve the robustness of neural networks. Prev...

Boosting Fast Adversarial Training with Learnable Adversarial Initialization

Adversarial training (AT) has been demonstrated to be effective in impro...

Improved Adversarial Training via Learned Optimizer

Adversarial attack has recently become a tremendous threat to deep learn...

Towards Improving Adversarial Training of NLP Models

Adversarial training, a method for learning robust deep neural networks,...

Adversarial Training Over Long-Tailed Distribution

In this paper, we study adversarial training on datasets that obey the l...

Adversarial Training with Complementary Labels: On the Benefit of Gradually Informative Attacks

Adversarial training (AT) with imperfect supervision is significant but ...

Fashion-Guided Adversarial Attack on Person Segmentation

This paper presents the first adversarial example based method for attac...

Please sign up or login with your details

Forgot password? Click here to reset