paper-with-me

홈 › Papers

Tight Robustness Certification Through the Convex Hull of $\ell_0$ Attacks

2025-11-13 · Yuval Shapira, Dana Drachsler-Cohen arxiv

Few-pixel attacks mislead a classifier by modifying a few pixels of an image. Their perturbation space is an $\ell_0$-ball, which is not convex, unlike $\ell_p$-balls for $p\geq1$. However, existing local robustness verifiers typically scale by relying on linear bound propagation, which captures convex perturbation spaces. We show that the convex hull of an $\ell_0$-ball is the intersection of its bounding box and an asymmetrically scaled $\ell_1$-like polytope. The volumes of the convex hull and this polytope are nearly equal as the input dimension increases. We then show a linear bound propagation that precisely computes bounds over the convex hull and is significantly tighter than bound propagations over the bounding box or our $\ell_1$-like polytope. This bound propagation scales the state-of-the-art $\ell_0$ verifier on its most challenging robustness benchmarks by 1.24x-7.07x, with a geometric mean of 3.16.

📄 PDF Abstract BibTeX arXiv:2511.10576

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Tightened Convex Relaxations for Neural Network Robustness Certification

2020-04-01 · Brendon G. Anderson, Ziye Ma, Jingqi Li, Somayeh Sojoudi

In this paper, we consider the problem of certifying the robustness of neural networks to perturbed and adversarial input data. Such certification is imperative for the application of neural networks in safety-critical d…

Decision Making

Improved Training of Certifiably Robust Models

2019-09-25 · Chen Zhu, Renkun Ni, Ping-Yeh Chiang, Hengduo Li 외

Convex relaxations are effective for training and certifying neural networks against norm-bounded adversarial attacks, but they leave a large gap between certifiable and empirical (PGD) robustness. In principle, relaxati…

Partition-based formulations for mixed-integer optimization of trained ReLU neural networks

2021-02-08 · NeurIPS 2021 12 · Calvin Tsay, Jan Kronqvist, Alexander Thebelt, Ruth Misener

This paper introduces a class of mixed-integer formulations for trained ReLU neural networks. The approach balances model size and tightness by partitioning node inputs into a number of groups and forming the convex hull…

On the tightness of linear relaxation based robustness certification methods

2022-10-01 · Cheng Tang

There has been a rapid development and interest in adversarial training and defenses in the machine learning community in the recent years. One line of research focuses on improving the performance and efficiency of adve…

Adversarial Robustness

Lipschitz-Based Robustness Certification for Recurrent Neural Networks via Convex Relaxation

2025-09-22 · Paul Hamelbeck, Johannes Schiffer arxiv

Robustness certification against bounded input noise or adversarial perturbations is increasingly important for deployment recurrent neural networks (RNNs) in safety-critical control applications. To address this challen…