• Home
  • Search
  • Adversarial example generation with adabelief optimizer and crop invariance
  • Open Access IconOpen Access
  • Cite Icon33
  • https://doi.org/10.1007/s10489-022-03469-5Copy DOI Icon

Adversarial example generation with adabelief optimizer and crop invariance

Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Deep neural networks are vulnerable to adversarial examples, which are crafted by applying small, human-imperceptible perturbations on the original images, so as to mislead deep neural networks to output inaccurate predictions. Adversarial attacks can thus be an important method to evaluate and select robust models in safety-critical applications. However, under the challenging black-box setting, most existing adversarial attacks often achieve relatively low success rates on normally trained networks and adversarially trained networks. In this paper, we regard the generation process of adversarial examples as an optimization process similar to deep neural network training. From this point of view, we introduce AdaBelief optimizer and crop invariance into the generation of adversarial examples, and propose AdaBelief Iterative Fast Gradient Method (ABI-FGM) and Crop-Invariant attack Method (CIM) to improve the transferability of adversarial examples. By adopting adaptive learning rate into the iterative attacks, ABI-FGM can optimize the convergence process, resulting in more transferable adversarial examples. CIM is based on our discovery on the crop-invariant property of deep neural networks, which we thus leverage to optimize the adversarial perturbations over an ensemble of crop copies so as to avoid overfitting on the white-box model being attacked and improve the transferability of adversarial examples. ABI-FGM and CIM can be readily integrated to build a strong gradient-based attack to further boost the success rates of adversarial examples for black-box attacks. Moreover, our method can also be naturally combined with other gradient-based attack methods to build a more robust attack to generate more transferable adversarial examples against the defense models. Extensive experiments on the ImageNet dataset demonstrate the method’s effectiveness. Whether on normally trained networks or adversarially trained networks, our method has higher success rates than state-of-the-art gradient-based attack methods.

Similar Papers
  • Research Article
  • Citations6

Strengthening transferability of adversarial examples by adaptive inertia and amplitude spectrum dropout

  • Jun 30, 2023
  • Neural Networks
  • Huanhuan Li +2
  • Conference Article
  • Citations16

Strengthening the Transferability of Adversarial Examples Using Advanced Looking Ahead and Self-CutMix

  • Jun 01, 2022
  • Donggon Jang +2
  • Research Article
  • Citations18

Frequency domain regularization for iterative adversarial attacks

  • Oct 05, 2022
  • Pattern Recognition
  • Tengjiao Li +3
  • Conference Article
  • Citations1021

Evading Defenses to Transferable Adversarial Examples by Translation-Invariant Attacks

  • Jun 01, 2019
  • Yinpeng Dong +3
  • Research Article

Transfer-based Adversarial Attack with Rectified Adam and Color Invariance

  • Jan 01, 2022
  • International Journal of Software and Informatics
  • Jia Ding +1
  • Research Article
  • Citations11

Direction-aggregated Attack for Transferable Adversarial Examples

  • Apr 28, 2022
  • ACM Journal on Emerging Technologies in Computing Systems
  • Tianjin Huang +4
  • Research Article
  • Citations6

Adversarial attack via dual-stage network erosion

  • Aug 19, 2022
  • Computers & Security
  • Yexin Duan +7
  • Conference Article
  • Citations3

Boosting Transferability of Adversarial Example via an Enhanced Euler’s Method

  • Jun 04, 2023
  • Anjie Peng +4
  • Research Article
  • Citations20

Toward Transferable Attack via Adversarial Diffusion in Face Recognition

  • Jan 01, 2024
  • IEEE Transactions on Information Forensics and Security
  • Cong Hu +3
  • Book Chapter

Learning Indistinguishable and Transferable Adversarial Examples

  • Jan 01, 2021
  • Wu Zhang +4
  • Book Chapter

Towards Transferable Adversarial Examples Using Meta Learning

  • Jan 01, 2022
  • Mingyuan Fan +3
  • Research Article
  • Citations6

Foolmix: Strengthen the Transferability of Adversarial Examples by Dual-Blending and Direction Update Strategy

  • Jan 01, 2024
  • IEEE Transactions on Information Forensics and Security
  • Zhankai Li +4
  • Research Article
  • Citations6

Enhancing transferability of adversarial examples via rotation‐invariant attacks

  • May 18, 2021
  • IET Computer Vision
  • Yexin Duan +5
  • Research Article
  • Citations5

Improving transferability of adversarial examples by saliency distribution and data augmentation

  • Jun 23, 2022
  • Computers & Security
  • Yansong Dong +4
  • Research Article
  • Citations3

Rethinking the optimization objective for transferable adversarial examples from a fuzzy perspective.

  • Apr 01, 2025
  • Neural networks : the official journal of the International Neural Network Society
  • Xiangyuan Yang +3
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.