Saved in:
Bibliographic Details
Main Authors: Zhao, Mengnan, Zhang, Lihe, Ye, Jingwen, Lu, Huchuan, Yin, Baocai, Wang, Xinchao
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2410.15042
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909356478431232
author Zhao, Mengnan
Zhang, Lihe
Ye, Jingwen
Lu, Huchuan
Yin, Baocai
Wang, Xinchao
author_facet Zhao, Mengnan
Zhang, Lihe
Ye, Jingwen
Lu, Huchuan
Yin, Baocai
Wang, Xinchao
contents Adversarial training (AT) refers to integrating adversarial examples -- inputs altered with imperceptible perturbations that can significantly impact model predictions -- into the training process. Recent studies have demonstrated the effectiveness of AT in improving the robustness of deep neural networks against diverse adversarial attacks. However, a comprehensive overview of these developments is still missing. This survey addresses this gap by reviewing a broad range of recent and representative studies. Specifically, we first describe the implementation procedures and practical applications of AT, followed by a comprehensive review of AT techniques from three perspectives: data enhancement, network design, and training configurations. Lastly, we discuss common challenges in AT and propose several promising directions for future research.
format Preprint
id arxiv_https___arxiv_org_abs_2410_15042
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Adversarial Training: A Survey
Zhao, Mengnan
Zhang, Lihe
Ye, Jingwen
Lu, Huchuan
Yin, Baocai
Wang, Xinchao
Machine Learning
Artificial Intelligence
Adversarial training (AT) refers to integrating adversarial examples -- inputs altered with imperceptible perturbations that can significantly impact model predictions -- into the training process. Recent studies have demonstrated the effectiveness of AT in improving the robustness of deep neural networks against diverse adversarial attacks. However, a comprehensive overview of these developments is still missing. This survey addresses this gap by reviewing a broad range of recent and representative studies. Specifically, we first describe the implementation procedures and practical applications of AT, followed by a comprehensive review of AT techniques from three perspectives: data enhancement, network design, and training configurations. Lastly, we discuss common challenges in AT and propose several promising directions for future research.
title Adversarial Training: A Survey
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2410.15042