Enhancing generalization in high-energy physics using white-box adversarial attacks

Article Article scientifique Article Enhancing generalization in high-energy physics using white-box adversarial attacks eng 2025-07-07 author Rothen Franck Alexis 0009-0009-1860-6581 Université de Genève 942294 Département de physique nucléaire et corpusculaire (FINUC) author Klein Samuel 0000-0002-2999-6150 Université de Genève 951964 author Leigh Matthew 0000-0003-1406-1413 Université de Genève 951970 author Golling Tobias 0000-0001-8535-6687 Université de Genève 827102 Département de physique nucléaire et corpusculaire (FINUC) 10.1103/PhysRevD.112.016004 2470-0010 https://link.aps.org/doi/10.1103/PhysRevD.112.016004 <p>Machine learning is becoming increasingly popular in the context of particle physics. Supervised learning, which uses labeled Monte Carlo (MC) simulations, remains one of the most widely used methods for discriminating signals beyond the Standard Model. However, this paper suggests that supervised models may depend excessively on artifacts and approximations from Monte Carlo simulations, potentially limiting their ability to generalize well to real data. This study aims to enhance the generalization properties of supervised models by reducing the sharpness of local minima. It reviews the application of four distinct white-box adversarial attacks in the context of classifying Higgs boson decay signals. The attacks are divided into weight-space attacks and feature-space attacks. To study and quantify the sharpness of different local minima, this paper presents two analysis methods: gradient ascent and reduced Hessian eigenvalue analysis. The results show that white-box adversarial attacks significantly improve generalization performance, albeit with increased computational complexity.</p> Dewey 500.2 Swiss National Science Foundation Robust Deep Density Models for High-Energy Particle Physics and Solar Flare Analysis (RODEM) 193716 CH 34 Département de physique nucléaire et corpusculaire 132 false Physical review. D 112 1 631eb3c6-c329-4397-8f6e-d6bfad4c0829 Article (Published version) Public Article_scientifique_from_crossref.pdf 1556694 application/pdf CC BY-4.0 2959ee7b4054e5b2fe22aa263d24a28cc6dd074b9057d1ae85517d1d97cd7e7a openAccess Faculté des sciences / Section de physique / Département de physique nucléaire et corpusculaire

Title	Published in	Access level	OA Policy	Year	Views	Downloads

Enhancing generalization in high-energy physics using white-box adversarial attacks	Physical review. D			2025	6	22

Archive ouverte UNIGE

Rothen, Franck

account_balance Affiliation entities

Research groups

Affiliation entities