• Home
  • Search
  • AROID: Improving Adversarial Robustness Through Online Instance-Wise Data Augmentation
  • Cite Icon7
  • https://doi.org/10.1007/s11263-024-02206-4Copy DOI Icon

AROID: Improving Adversarial Robustness Through Online Instance-Wise Data Augmentation

Show More
  • Abstract
  • Highlights & Summary
  • PDF
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Deep neural networks are vulnerable to adversarial examples. Adversarial training (AT) is an effective defense against adversarial examples. However, AT is prone to overfitting which degrades robustness substantially. Recently, data augmentation (DA) was shown to be effective in mitigating robust overfitting if appropriately designed and optimized for AT. This work proposes a new method to automatically learn online, instance-wise, DA policies to improve robust generalization for AT. This is the first automated DA method specific for robustness. A novel policy learning objective, consisting of Vulnerability, Affinity and Diversity, is proposed and shown to be sufficiently effective and efficient to be practical for automatic DA generation during AT. Importantly, our method dramatically reduces the cost of policy search from the 5000 h of AutoAugment and the 412 h of IDBH to 9 h, making automated DA more practical to use for adversarial robustness. This allows our method to efficiently explore a large search space for a more effective DA policy and evolve the policy as training progresses. Empirically, our method is shown to outperform all competitive DA methods across various model architectures and datasets. Our DA policy reinforced vanilla AT to surpass several state-of-the-art AT methods regarding both accuracy and robustness. It can also be combined with those advanced AT methods to further boost robustness. Code and pre-trained models are available at: https://github.com/TreeLLi/AROID.

Loading PDF

Similar Papers
  • PDF
  • Research Article
  • Citations9

LADDER: Latent boundary-guided adversarial training

  • Sep 09, 2022
  • Machine Learning
  • Xiaowei Zhou +2
  • Research Article
  • Citations9

F2AT: Feature-Focusing Adversarial Training via Disentanglement of Natural and Perturbed Patterns

  • Sep 01, 2025
  • IEEE Transactions on Knowledge and Data Engineering
  • Yaguan Qian +6
  • Conference Article

A General Multiple Data Augmentation Based Framework for Training Deep Neural Networks

  • Jul 18, 2022
  • Binyan Hu +2
  • Research Article
  • Citations5

Toward deep neural networks robust to adversarial examples, using augmented data importance perception

  • Dec 13, 2022
  • Journal of Electronic Imaging
  • Zhiming Chen +4
  • Research Article
  • Citations14

Increasing-Margin Adversarial (IMA) training to improve adversarial robustness of neural networks

  • Jun 24, 2023
  • Computer Methods and Programs in Biomedicine
  • Linhai Ma +1
  • Book Chapter

Weighted Adaptive Perturbations Adversarial Training for Improving Robustness

  • Jan 01, 2022
  • Yan Wang +2
  • PDF
  • Research Article
  • Citations3

Between-Class Adversarial Training for Improving Adversarial Robustness of Image Classification.

  • Mar 20, 2023
  • Sensors (Basel, Switzerland)
  • Desheng Wang +2
  • Book Chapter
  • Citations81

Adversarial Training Against Location-Optimized Adversarial Patches

  • Jan 01, 2020
  • Sukrut Rao +2
  • Research Article
  • Citations4

A Comprehensive Understanding of the Impact of Data Augmentation on the Transferability of 3D Adversarial Examples

  • Feb 14, 2025
  • ACM Transactions on Knowledge Discovery from Data
  • Fulan Qian +6
  • Conference Article
  • Citations5

Enhancing Adversarial Robustness For Image Classification By Regularizing Class Level Feature Distribution

  • Sep 19, 2021
  • Cheng Yu +4
  • Conference Article
  • Citations7

CgAT: Center-Guided Adversarial Training for Deep Hashing-Based Retrieval

  • Apr 30, 2023
  • Xunguang Wang +2
  • Research Article
  • Citations28

Boosting adversarial robustness via self-paced adversarial training

  • Sep 09, 2023
  • Neural Networks
  • Lirong He +5
  • Research Article
  • Citations9

Towards robust neural networks via orthogonal diversity

  • Jan 12, 2024
  • Pattern Recognition
  • Kun Fang +7
  • Conference Article
  • Citations1

Robustness through Cognitive Dissociation Mitigation in Contrastive Adversarial Training

  • Sep 09, 2022
  • Adir Rahamim +1
  • Research Article
  • Citations11

Improving adversarial robustness using knowledge distillation guided by attention information bottleneck

  • Mar 05, 2024
  • Information Sciences
  • Yuxin Gong +4
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.