Minimax risk classifiers with 0-1 loss

TL;DR

This paper introduces minimax risk classifiers that optimize worst-case 0-1 loss over distribution sets, offering strong guarantees and practical optimization methods for accurate classification.

Contribution

It presents a novel minimax risk classifier framework that minimizes worst-case 0-1 loss with universal consistency and efficient optimization techniques.

Findings

MRCs provide tight performance guarantees.

MRCs are strongly universally consistent.

Methods achieve accurate classification with theoretical guarantees.

Abstract

Supervised classification techniques use training samples to learn a classification rule with small expected 0-1 loss (error probability). Conventional methods enable tractable learning and provide out-of-sample generalization by using surrogate losses instead of the 0-1 loss and considering specific families of rules (hypothesis classes). This paper presents minimax risk classifiers (MRCs) that minize the worst-case 0-1 loss with respect to uncertainty sets of distributions that can include the underlying distribution, with a tunable confidence. We show that MRCs can provide tight performance guarantees at learning and are strongly universally consistent using feature mappings given by characteristic kernels. The paper also proposes efficient optimization techniques for MRC learning and shows that the methods presented can provide accurate classification together with tight performance guarantees in practice.