A Closer Look at the Adversarial Robustness of Information Bottleneck Models
We study the adversarial robustness of information bottleneck models for classification. Previous works showed that the robustness of models trained with information bottlenecks can improve upon adversarial training. Our evaluation under a diverse range of white-box $l_{\infty}$ attacks suggests that information bottlenecks alone are not a strong defense strategy, and that previous results were likely influenced by gradient obfuscation.
Code (0)
등록된 구현이 없습니다.
Tasks
Adversarial RobustnessSimilar Papers 제목 키워드 기반
A Closer Look into the Robustness of Neural Dependency Parsers Using Better Adversarial Examples
MIMIR: Masked Image Modeling for Mutual Information-based Adversarial Robustness
Vision Transformers (ViTs) achieve excellent performance in various tasks, but they are also vulnerable to adversarial attacks. Building robust ViTs is highly dependent on dedicated Adversarial Training (AT) strategies. …
Adversarial RobustnessRevisiting Hilbert-Schmidt Information Bottleneck for Adversarial Robustness
We investigate the HSIC (Hilbert-Schmidt independence criterion) bottleneck as a regularizer for learning an adversarially robust deep neural network classifier. In addition to the usual cross-entropy loss, we add regula…
Adversarial RobustnessRobustness and Regularization in Hierarchical Re-Basin
This paper takes a closer look at Git Re-Basin, an interesting new approach to merge trained models. We propose a hierarchical model merging scheme that significantly outperforms the standard MergeMany algorithm. With ou…
Improving Adversarial Robustness via Information Bottleneck Distillation
Previous studies have shown that optimizing the information bottleneck can significantly improve the robustness of deep neural networks. Our study closely examines the information bottleneck principle and proposes an Inf…