Blending adversarial training and representation-conditional purification via aggregation improves adversarial robustness

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ballarin, Emanuele, Ansuini, Alessio, Bortolussi, Luca
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915532796592128
author Ballarin, Emanuele
Ansuini, Alessio
Bortolussi, Luca
author_facet Ballarin, Emanuele
Ansuini, Alessio
Bortolussi, Luca
contents In this work, we propose a novel adversarial defence mechanism for image classification - CARSO - blending the paradigms of adversarial training and adversarial purification in a synergistic robustness-enhancing way. The method builds upon an adversarially-trained classifier, and learns to map its internal representation associated with a potentially perturbed input onto a distribution of tentative clean reconstructions. Multiple samples from such distribution are classified by the same adversarially-trained model, and a carefully chosen aggregation of its outputs finally constitutes the robust prediction of interest. Experimental evaluation by a well-established benchmark of strong adaptive attacks, across different image datasets, shows that CARSO is able to defend itself against adaptive end-to-end white-box attacks devised for stochastic defences. Paying a modest clean accuracy toll, our method improves by a significant margin the state-of-the-art for Cifar-10, Cifar-100, and TinyImageNet-200 $\ell_\infty$ robust classification accuracy against AutoAttack. Code, and instructions to obtain pre-trained models are available at: https://github.com/emaballarin/CARSO .
format Preprint
id arxiv_https___arxiv_org_abs_2306_06081
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Blending adversarial training and representation-conditional purification via aggregation improves adversarial robustness
Ballarin, Emanuele
Ansuini, Alessio
Bortolussi, Luca
Computer Vision and Pattern Recognition
Artificial Intelligence
Cryptography and Security
Machine Learning
In this work, we propose a novel adversarial defence mechanism for image classification - CARSO - blending the paradigms of adversarial training and adversarial purification in a synergistic robustness-enhancing way. The method builds upon an adversarially-trained classifier, and learns to map its internal representation associated with a potentially perturbed input onto a distribution of tentative clean reconstructions. Multiple samples from such distribution are classified by the same adversarially-trained model, and a carefully chosen aggregation of its outputs finally constitutes the robust prediction of interest. Experimental evaluation by a well-established benchmark of strong adaptive attacks, across different image datasets, shows that CARSO is able to defend itself against adaptive end-to-end white-box attacks devised for stochastic defences. Paying a modest clean accuracy toll, our method improves by a significant margin the state-of-the-art for Cifar-10, Cifar-100, and TinyImageNet-200 $\ell_\infty$ robust classification accuracy against AutoAttack. Code, and instructions to obtain pre-trained models are available at: https://github.com/emaballarin/CARSO .
title Blending adversarial training and representation-conditional purification via aggregation improves adversarial robustness
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Cryptography and Security
Machine Learning
url https://arxiv.org/abs/2306.06081