Transferability Bound Theory: Exploring Relationship between Adversarial Transferability and Flatness

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Fan, Mingyuan, Li, Xiaodan, Chen, Cen, Zhou, Wenmeng, Li, Yaliang
Formato: Preprint
Publicado: 2023
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866909340861988864
author Fan, Mingyuan
Li, Xiaodan
Chen, Cen
Zhou, Wenmeng
Li, Yaliang
author_facet Fan, Mingyuan
Li, Xiaodan
Chen, Cen
Zhou, Wenmeng
Li, Yaliang
contents A prevailing belief in attack and defense community is that the higher flatness of adversarial examples enables their better cross-model transferability, leading to a growing interest in employing sharpness-aware minimization and its variants. However, the theoretical relationship between the transferability of adversarial examples and their flatness has not been well established, making the belief questionable. To bridge this gap, we embark on a theoretical investigation and, for the first time, derive a theoretical bound for the transferability of adversarial examples with few practical assumptions. Our analysis challenges this belief by demonstrating that the increased flatness of adversarial examples does not necessarily guarantee improved transferability. Moreover, building upon the theoretical analysis, we propose TPA, a Theoretically Provable Attack that optimizes a surrogate of the derived bound to craft adversarial examples. Extensive experiments across widely used benchmark datasets and various real-world applications show that TPA can craft more transferable adversarial examples compared to state-of-the-art baselines. We hope that these results can recalibrate preconceived impressions within the community and facilitate the development of stronger adversarial attack and defense mechanisms. The source codes are available in <https://github.com/fmy266/TPA>.
format Preprint
id arxiv_https___arxiv_org_abs_2311_06423
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Transferability Bound Theory: Exploring Relationship between Adversarial Transferability and Flatness
Fan, Mingyuan
Li, Xiaodan
Chen, Cen
Zhou, Wenmeng
Li, Yaliang
Machine Learning
Cryptography and Security
Computer Vision and Pattern Recognition
A prevailing belief in attack and defense community is that the higher flatness of adversarial examples enables their better cross-model transferability, leading to a growing interest in employing sharpness-aware minimization and its variants. However, the theoretical relationship between the transferability of adversarial examples and their flatness has not been well established, making the belief questionable. To bridge this gap, we embark on a theoretical investigation and, for the first time, derive a theoretical bound for the transferability of adversarial examples with few practical assumptions. Our analysis challenges this belief by demonstrating that the increased flatness of adversarial examples does not necessarily guarantee improved transferability. Moreover, building upon the theoretical analysis, we propose TPA, a Theoretically Provable Attack that optimizes a surrogate of the derived bound to craft adversarial examples. Extensive experiments across widely used benchmark datasets and various real-world applications show that TPA can craft more transferable adversarial examples compared to state-of-the-art baselines. We hope that these results can recalibrate preconceived impressions within the community and facilitate the development of stronger adversarial attack and defense mechanisms. The source codes are available in <https://github.com/fmy266/TPA>.
title Transferability Bound Theory: Exploring Relationship between Adversarial Transferability and Flatness
topic Machine Learning
Cryptography and Security
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2311.06423