Nonzero-sum adversarial hypothesis testing games

Yasodharan, S and Loiseau, P (2019) Nonzero-sum adversarial hypothesis testing games. In: 33rd Annual Conference on Neural Information Processing Systems, NeurIPS 2019, 8-14 December 2019, Vancouver; Canada.

PDF adv_neu_inf_pro_sys_32_2019.pdf - Published Version Restricted to Registered users only Download (694kB) | Request a copy

Abstract

We study nonzero-sum hypothesis testing games that arise in the context of adversarial classification, in both the Bayesian as well as the Neyman-Pearson frameworks. We first show that these games admit mixed strategy Nash equilibria, and then we examine some interesting concentration phenomena of these equilibria. Our main results are on the exponential rates of convergence of classification errors at equilibrium, which are analogous to the well-known Chernoff-Stein lemma and Chernoff information that describe the error exponents in the classical binary hypothesis testing problem, but with parameters derived from the adversarial model. The results are validated through numerical experiments. © 2019 Neural information processing systems foundation. All rights reserved.

Item Type:	Conference Paper
Publication:	Advances in Neural Information Processing Systems
Publisher:	Neural information processing systems foundation
Additional Information:	The copyright of this article belongs to Neural information processing systems foundation
Keywords:	Classification (of information); Statistical tests, Adversarial classifications; Binary Hypothesis Testing; Chernoff information; Classification errors; Concentration phenomena; Exponential rates; Hypothesis testing; Numerical experiments, Well testing
Department/Centre:	Division of Electrical Sciences > Electrical Communication Engineering
Date Deposited:	22 Sep 2020 07:25
Last Modified:	29 Aug 2022 10:51
URI:	https://eprints.iisc.ac.in/id/eprint/66570

Actions (login required)

View Item