26
7

Bayesian Robustness: A Nonasymptotic Viewpoint

Abstract

We study the problem of robustly estimating the posterior distribution for the setting where observed data can be contaminated with potentially adversarial outliers. We propose Rob-ULA, a robust variant of the Unadjusted Langevin Algorithm (ULA), and provide a finite-sample analysis of its sampling distribution. In particular, we show that after T=O~(d/εacc)T= \tilde{\mathcal{O}}(d/\varepsilon_{\textsf{acc}}) iterations, we can sample from pTp_T such that dist(pT,p)εacc+O~(ϵ)\text{dist}(p_T, p^*) \leq \varepsilon_{\textsf{acc}} + \tilde{\mathcal{O}}(\epsilon), where ϵ\epsilon is the fraction of corruptions. We corroborate our theoretical analysis with experiments on both synthetic and real-world data sets for mean estimation, regression and binary classification.

View on arXiv
Comments on this paper