DeepAI AI Chat
Log In Sign Up

Strength-Adaptive Adversarial Training

10/04/2022
by   Chaojian Yu, et al.
The University of Sydney
Hong Kong Baptist University
The University of Melbourne
Xidian University
USTC
0

Adversarial training (AT) is proved to reliably improve network's robustness against adversarial data. However, current AT with a pre-specified perturbation budget has limitations in learning a robust network. Firstly, applying a pre-specified perturbation budget on networks of various model capacities will yield divergent degree of robustness disparity between natural and robust accuracies, which deviates from robust network's desideratum. Secondly, the attack strength of adversarial training data constrained by the pre-specified perturbation budget fails to upgrade as the growth of network robustness, which leads to robust overfitting and further degrades the adversarial robustness. To overcome these limitations, we propose Strength-Adaptive Adversarial Training (SAAT). Specifically, the adversary employs an adversarial loss constraint to generate adversarial training data. Under this constraint, the perturbation budget will be adaptively adjusted according to the training state of adversarial data, which can effectively avoid robust overfitting. Besides, SAAT explicitly constrains the attack strength of training data through the adversarial loss, which manipulates model capacity scheduling during training, and thereby can flexibly control the degree of robustness disparity and adjust the tradeoff between natural accuracy and robustness. Extensive experiments show that our proposal boosts the robustness of adversarial training.

READ FULL TEXT

page 1

page 2

page 3

page 4

05/30/2022

Robust Weight Perturbation for Adversarial Training

Overfitting widely exists in adversarial robust training of deep network...
12/03/2018

Measuring the Robustness of Graph Properties

In this paper, we propose a perturbation framework to measure the robust...
06/17/2022

Understanding Robust Overfitting of Adversarial Training and Beyond

Robust overfitting widely exists in adversarial training of deep network...
10/15/2020

Overfitting or Underfitting? Understand Robustness Drop in Adversarial Training

Our goal is to understand why the robustness drops after conducting adve...
05/09/2019

Exploring the Hyperparameter Landscape of Adversarial Robustness

Adversarial training shows promise as an approach for training models th...
03/18/2020

Improving Adversarial Robustness Through Progressive Hardening

Adversarial training (AT) has become a popular choice for training robus...