paper-with-me

홈 › Papers

Tight Second-Order Certificates for Randomized Smoothing

2020-10-20 · Alexander Levine, Aounon Kumar, Thomas Goldstein, Soheil Feizi

Randomized smoothing is a popular way of providing robustness guarantees against adversarial attacks: randomly-smoothed functions have a universal Lipschitz-like bound, allowing for robustness certificates to be easily computed. In this work, we show that there also exists a universal curvature-like bound for Gaussian random smoothing: given the exact value and gradient of a smoothed function, we compute a lower bound on the distance of a point to its closest adversarial example, called the Second-order Smoothing (SoS) robustness certificate. In addition to proving the correctness of this novel certificate, we show that SoS certificates are realizable and therefore tight. Interestingly, we show that the maximum achievable benefits, in terms of certified robustness, from using the additional information of the gradient norm are relatively small: because our bounds are tight, this is a fundamental negative result. The gain of SoS certificates further diminishes if we consider the estimation error of the gradient norms, for which we have developed an estimator. We therefore additionally develop a variant of Gaussian smoothing, called Gaussian dipole smoothing, which provides similar bounds to randomized smoothing with gradient information, but with much-improved sample efficiency. This allows us to achieve (marginally) improved robustness certificates on high-dimensional datasets such as CIFAR-10 and ImageNet. Code is available at https://github.com/alevine0/smoothing_second_order.

📄 PDF Abstract BibTeX arXiv:2010.10549

Code (1)

alevine0/smoothing_second_order 공식 구현 pytorch

Methods 이 논문이 사용한 방법론

Randomized Smoothing 설명 없음

Similar Papers 제목 키워드 기반

Invariance-Aware Randomized Smoothing Certificates

2022-11-25 · Jan Schuchardt, Stephan Günnemann

Building models that comply with the invariances inherent to different domains, such as invariance under translation or rotation, is a key aspect of applying machine learning to real world problems like molecular propert…

Molecular Property PredictionProperty PredictionProtein Folding

Sound Randomized Smoothing in Floating-Point Arithmetics

2022-07-14 · Václav Voráček, Matthias Hein

Randomized smoothing is sound when using infinite precision. However, we show that randomized smoothing is no longer sound for limited floating-point precision. We present a simple example where randomized smoothing cert…

image-classificationImage Classification

Higher-Order Certified Robustness for Regression

2026-07-06 · Jie Zhang, Natalie Frank arxiv

Randomized smoothing has emerged as a scalable technique for certifying the adversarial robustness of classifiers. However, its application to regression remains under-explored and faces unique challenges. Existing regre…

Adversarial Robustness

$\ell_1$ Adversarial Robustness Certificates: a Randomized Smoothing Approach

2019-09-25 · Jiaye Teng, Guang-He Lee, Yang Yuan

Robustness is an important property to guarantee the security of machine learning models. It has recently been demonstrated that strong robustness certificates can be obtained on ensemble classifiers generated by input r…

Adversarial Robustness

Certified Robustness under Heterogeneous Perturbations via Hybrid Randomized Smoothing

2026-05-13 · Blaise Delattre, Hengyu Wu, Paul Caillon, Wei Yang Bryan Lim 외 arxiv

Randomized smoothing provides strong, model-agnostic robustness certificates, but existing guarantees are limited to single modalities, treating continuous and discrete inputs in isolation. This limitation becomes critic…