paper-with-me

Papers

Universal Adversarial Training with Class-Wise Perturbations

2021-04-07 · Philipp Benz, Chaoning Zhang, Adil Karjauv, In So Kweon

Despite their overwhelming success on a wide range of applications, convolutional neural networks (CNNs) are widely recognized to be vulnerable to adversarial examples. This intriguing phenomenon led to a competition between adversarial attacks and defense techniques. So far, adversarial training is the most widely used method for defending against adversarial attacks. It has also been extended to defend against universal adversarial perturbations (UAPs). The SOTA universal adversarial training (UAT) method optimizes a single perturbation for all training samples in the mini-batch. In this work, we find that a UAP does not attack all classes equally. Inspired by this observation, we identify it as the source of the model having unbalanced robustness. To this end, we improve the SOTA UAT by proposing to utilize class-wise UAPs during adversarial training. On multiple benchmark datasets, our class-wise UAT leads superior performance for both clean accuracy and adversarial robustness against universal attack.

📄 PDF Abstract BibTeX arXiv:2104.03000

Code (0)

등록된 구현이 없습니다.

Tasks

Adversarial Robustness

Similar Papers 제목 키워드 기반

A Method for Computing Class-wise Universal Adversarial Perturbations

2019-12-01 · Tejus Gupta, Abhishek Sinha, Nupur Kumari, Mayank Singh 외

We present an algorithm for computing class-specific universal adversarial perturbations for deep neural networks. Such perturbations can induce misclassification in a large fraction of images of a specific class. Unlike…

Defending Against Universal Perturbations With Shared Adversarial Training

2018-12-10 · ICCV 2019 10 · Chaithanya Kumar Mummadi, Thomas Brox, Jan Hendrik Metzen

Classifiers such as deep neural networks have been shown to be vulnerable against adversarial perturbations on problems with high-dimensional input space. While adversarial training improves the robustness of image class…

image-classificationImage ClassificationSemantic Segmentation

Universal Adversarial Perturbations Against Semantic Image Segmentation

2017-04-19 · ICCV 2017 10 · Jan Hendrik Metzen, Mummadi Chaithanya Kumar, Thomas Brox, Volker Fischer

While deep learning is remarkably successful on perceptual tasks, it was also shown to be vulnerable to adversarial perturbations of the input. These perturbations denote noise added to the input that was generated speci…

image-classificationImage ClassificationImage SegmentationSegmentation+1

Universal Adversarial Attack Using Very Few Test Examples

2019-09-25 · Amit Deshpande, Sandesh Kamath, K V Subrahmanyam

Adversarial attacks such as Gradient-based attacks, Fast Gradient Sign Method (FGSM) by Goodfellow et al.(2015) and DeepFool by Moosavi-Dezfooli et al. (2016) are input-dependent, small pixel-wise perturbations of images…

Adversarial Attack

Universal Adversarial Training

2018-11-27 · Ali Shafahi, Mahyar Najibi, Zheng Xu, John Dickerson 외

Standard adversarial attacks change the predicted class label of a selected image by adding specially tailored small perturbations to its pixels. In contrast, a universal perturbation is an update that can be added to an…