Adversarial classification via distributional robustness with Wasserstein ambiguity

We study a model for adversarial classification based on distributionally robust chance constraints. We show that under Wasserstein ambiguity, the model aims to minimize the conditional value-at-risk of the distance to misclassification, and we explore links to adversarial classification models prop...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Mathematical programming Ročník 198; číslo 2; s. 1411 - 1447
Hlavní autoři: Ho-Nguyen, Nam, Wright, Stephen J.
Médium: Journal Article
Jazyk:angličtina
Vydáno: Berlin/Heidelberg Springer Berlin Heidelberg 01.04.2023
Springer
Témata:
Calculus of Variations and Optimal Control; Optimization
Combinatorics
Full Length Paper
Mathematical and Computational Physics
Mathematical Methods in Physics
Mathematics
Mathematics and Statistics
Mathematics of Computing
Numerical Analysis
Theoretical
90C30
Wasserstein ambiguity
Ramp loss
68T09
90C17
Adversarial cassification
90C26
Nonconvex
Distributional robustness
Margin
ISSN:0025-5610, 1436-4646
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:We study a model for adversarial classification based on distributionally robust chance constraints. We show that under Wasserstein ambiguity, the model aims to minimize the conditional value-at-risk of the distance to misclassification, and we explore links to adversarial classification models proposed earlier and to maximum-margin classifiers. We also provide a reformulation of the distributionally robust model for linear classification, and show it is equivalent to minimizing a regularized ramp loss objective. Numerical experiments show that, despite the nonconvexity of this formulation, standard descent methods appear to converge to the global minimizer for this problem. Inspired by this observation, we show that, for a certain class of distributions, the only stationary point of the regularized ramp loss minimization problem is the global minimizer.
ISSN:0025-5610
1436-4646
DOI:10.1007/s10107-022-01796-6