On Improving Adversarial Transferability of Vision Transformers
arXiv:2106.04169
Abstract
Vision transformers (ViTs) process input images as sequences of patches via self-attention; a radically different architecture than convolutional neural networks (CNNs). This makes it interesting to study the adversarial feature space of ViT models and their transferability. In particular, we observe that adversarial patterns found via conventional adversarial attacks show very \emph{low} black-box transferability even for large ViT models. We show that this phenomenon is only due to the sub-optimal attack procedures that do not leverage the true representation potential of ViTs. A deep ViT is composed of multiple blocks, with a consistent architecture comprising of self-attention and feed-forward layers, where each block is capable of independently producing a class token. Formulating an attack using only the last class token (conventional approach) does not directly leverage the discriminative information stored in the earlier tokens, leading to poor adversarial transferability of ViTs. Using the compositional nature of ViT models, we enhance transferability of existing attacks by introducing two novel strategies specific to the architecture of ViT models. (i) Self-Ensemble: We propose a method to find multiple discriminative pathways by dissecting a single ViT model into an ensemble of networks. This allows explicitly utilizing class-specific information at each ViT block. (ii) Token Refinement: We then propose to refine the tokens to further enhance the discriminative capacity at each block of ViT. Our token refinement systematically combines the class tokens with structural information preserved within the patch tokens.
ICLR'22 (Spotlight), the first two authors contributed equally. Code: https://t.ly/hBbW
References in corpus (14)
- An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
- Transformers in Vision: A Survey
- Language Models are Few-Shot Learners
- ZOO: Zeroth Order Optimization based Black-box Attacks to Deep Neural Networks without Training Substitute Models
- MLP-Mixer: An all-MLP Architecture for Vision
- Reliable evaluation of adversarial robustness with an ensemble of diverse parameter-free attacks
- Skip Connections Matter: On the Transferability of Adversarial Examples Generated with ResNets
- A Study of Face Obfuscation in ImageNet
- Diversity can be Transferred: Output Diversification for White- and Black-box Attacks
- Perturbing Across the Feature Hierarchy to Improve Standard and Strict Blackbox Attack Transferability
- Enhancing the Transferability of Adversarial Attacks through Variance Tuning
- Are Convolutional Neural Networks or Transformers more like human vision?
- Transferable Perturbations of Deep Feature Distributions
- Backpropagating Linearly Improves Transferability of Adversarial Examples