paper-with-me

Papers

Exploring the Relationship between Architecture and Adversarially Robust Generalization

2022-09-28 · Aishan Liu, Shiyu Tang, Siyuan Liang, Ruihao Gong, Boxi Wu, Xianglong Liu, DaCheng Tao

Adversarial training has been demonstrated to be one of the most effective remedies for defending adversarial examples, yet it often suffers from the huge robustness generalization gap on unseen testing adversaries, deemed as the adversarially robust generalization problem. Despite the preliminary understandings devoted to adversarially robust generalization, little is known from the architectural perspective. To bridge the gap, this paper for the first time systematically investigated the relationship between adversarially robust generalization and architectural design. Inparticular, we comprehensively evaluated 20 most representative adversarially trained architectures on ImageNette and CIFAR-10 datasets towards multiple `p-norm adversarial attacks. Based on the extensive experiments, we found that, under aligned settings, Vision Transformers (e.g., PVT, CoAtNet) often yield better adversarially robust generalization while CNNs tend to overfit on specific attacks and fail to generalize on multiple adversaries. To better understand the nature behind it, we conduct theoretical analysis via the lens of Rademacher complexity. We revealed the fact that the higher weight sparsity contributes significantly towards the better adversarially robust generalization of Transformers, which can be often achieved by the specially-designed attention blocks. We hope our paper could help to better understand the mechanism for designing robust DNNs. Our model weights can be found at http://robust.art.

📄 PDF Abstract BibTeX arXiv:2209.14105

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Residual Connection 설명 없음
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Multi-Head Attention 설명 없음
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…

Similar Papers 제목 키워드 기반

Exploring the Relationship Between Architectural Design and Adversarially Robust Generalization

2023-01-01 · CVPR 2023 1 · Aishan Liu, Shiyu Tang, Siyuan Liang, Ruihao Gong 외

Adversarial training has been demonstrated to be one of the most effective remedies for defending adversarial examples, yet it often suffers from the huge robustness generalization gap on unseen testing adversaries, …

Improved Sample Complexities for Deep Networks and Robust Classification via an All-Layer Margin

2019-10-09 · Colin Wei, Tengyu Ma

For linear classifiers, the relationship between (normalized) output margin and generalization is captured in a clear and simple bound -- a large output margin implies good generalization. Unfortunately, for deep models,…

AllGeneral ClassificationGeneralization BoundsRobust classification

Improved Sample Complexities for Deep Neural Networks and Robust Classification via an All-Layer Margin

2020-05-01 · ICLR 2020 1 · Colin Wei, Tengyu Ma

For linear classifiers, the relationship between (normalized) output margin and generalization is captured in a clear and simple bound – a large output margin implies good generalization. Unfortunately, for deep models, …

AllGeneralization BoundsRobust classification

Robust Local Features for Improving the Generalization of Adversarial Training

2019-09-23 · ICLR 2020 1 · Chuanbiao Song, Kun He, Jiadong Lin, Li-Wei Wang 외

Adversarial training has been demonstrated as one of the most effective methods for training robust models to defend against adversarial examples. However, adversarially trained models often lack adversarially robust gen…

Understanding Adversarially Robust Generalization via Weight-Curvature Index

2024-10-10 · Yuelin Xu, Xiao Zhang

Despite extensive research on adversarial examples, the underlying mechanisms of adversarially robust generalization, a critical yet challenging task for deep learning, remain largely unknown. In this work, we propose a …

Adversarial RobustnessGeneralization Bounds