paper-with-me

Papers

Efficient Certified Defenses Against Patch Attacks on Image Classifiers

2021-02-08 · ICLR 2021 1 · Jan Hendrik Metzen, Maksym Yatsura

Adversarial patches pose a realistic threat model for physical world attacks on autonomous systems via their perception component. Autonomous systems in safety-critical domains such as automated driving should thus contain a fail-safe fallback component that combines certifiable robustness against patches with efficient inference while maintaining high performance on clean inputs. We propose BagCert, a novel combination of model architecture and certification procedure that allows efficient certification. We derive a loss that enables end-to-end optimization of certified robustness against patches of different sizes and locations. On CIFAR10, BagCert certifies 10.000 examples in 43 seconds on a single GPU and obtains 86% clean and 60% certified accuracy against 5x5 patches.

📄 PDF Abstract BibTeX arXiv:2102.04154

Code (0)

등록된 구현이 없습니다.

Tasks

GPU

Similar Papers 제목 키워드 기반

Certified Defenses for Adversarial Patches

2020-03-14 · ICLR 2020 1 · Ping-Yeh Chiang, Renkun Ni, Ahmed Abdelkader, Chen Zhu 외

Adversarial patch attacks are among one of the most practical threat models against real-world computer vision systems. This paper studies certified and empirical defenses against patch attacks. We begin with a set of ex…

ScaleCert: Scalable Certified Defense against Adversarial Patches with Sparse Superficial Layers

2021-10-27 · NeurIPS 2021 12 · Husheng Han, Kaidi Xu, Xing Hu, Xiaobing Chen 외

Adversarial patch attacks that craft the pixels in a confined region of the input images show their powerful attack effectiveness in physical environments even with noises or deformations. Existing certified defenses tow…

(De)Randomized Smoothing for Certifiable Defense against Patch Attacks

2020-02-25 · NeurIPS 2020 12 · Alexander Levine, Soheil Feizi

Patch adversarial attacks on images, in which the attacker can distort pixels within a region of bounded size, are an important threat model since they provide a quantitative model for physical adversarial attacks. In th…

Architecture-agnostic Iterative Black-box Certified Defense against Adversarial Patches

2023-05-18 · Di Yang, Yihao Huang, Qing Guo, Felix Juefei-Xu 외

The adversarial patch attack aims to fool image classifiers within a bounded, contiguous region of arbitrary changes, posing a real threat to computer vision systems (e.g., autonomous driving, content moderation, biometr…

Autonomous Driving

CertVLA: Certified Defense against Physical Visual Attacks for Vision-Language-Action Models

2026-08-21 · Hui Lu, Zhijie Peng, Yuqi Lin, Zaijia Yang 외 arxiv

Vision-Language-Action (VLA) policies are vulnerable to localized physical perturbations, yet existing certified patch defenses target discrete labels and cannot directly certify continuous, temporally correlated actions…