Salvato in:
Dettagli Bibliografici
Autori principali: Zeng, Hui, Chen, Biwei, Peng, Anjie
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2401.02727
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909072058482688
author Zeng, Hui
Chen, Biwei
Peng, Anjie
author_facet Zeng, Hui
Chen, Biwei
Peng, Anjie
contents Adversarial examples (AEs) have been extensively studied due to their potential for privacy protection and inspiring robust neural networks. Yet, making a targeted AE transferable across unknown models remains challenging. In this paper, to alleviate the overfitting dilemma common in an AE crafted by existing simple iterative attacks, we propose fine-tuning it in the feature space. Specifically, starting with an AE generated by a baseline attack, we encourage the features conducive to the target class and discourage the features to the original class in a middle layer of the source model. Extensive experiments demonstrate that only a few iterations of fine-tuning can boost existing attacks' targeted transferability nontrivially and universally. Our results also verify that the simple iterative attacks can yield comparable or even better transferability than the resource-intensive methods, which rest on training target-specific classifiers or generators with additional data. The code is available at: github.com/zengh5/TA_feature_FT.
format Preprint
id arxiv_https___arxiv_org_abs_2401_02727
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Enhancing targeted transferability via feature space fine-tuning
Zeng, Hui
Chen, Biwei
Peng, Anjie
Computer Vision and Pattern Recognition
Artificial Intelligence
Adversarial examples (AEs) have been extensively studied due to their potential for privacy protection and inspiring robust neural networks. Yet, making a targeted AE transferable across unknown models remains challenging. In this paper, to alleviate the overfitting dilemma common in an AE crafted by existing simple iterative attacks, we propose fine-tuning it in the feature space. Specifically, starting with an AE generated by a baseline attack, we encourage the features conducive to the target class and discourage the features to the original class in a middle layer of the source model. Extensive experiments demonstrate that only a few iterations of fine-tuning can boost existing attacks' targeted transferability nontrivially and universally. Our results also verify that the simple iterative attacks can yield comparable or even better transferability than the resource-intensive methods, which rest on training target-specific classifiers or generators with additional data. The code is available at: github.com/zengh5/TA_feature_FT.
title Enhancing targeted transferability via feature space fine-tuning
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2401.02727