MRCpy: A Library for Minimax Risk Classifiers

TL;DR

MRCpy is a Python library that implements minimax risk classifiers using the robust risk minimization approach, offering performance guarantees and adaptability to distribution shifts, with seamless integration into existing Python ML workflows.

Contribution

It introduces MRCpy, a new library for minimax risk classifiers based on RRM, expanding the toolkit beyond traditional ERM-based methods.

Findings

Provides multiple variants of MRCs with performance guarantees

Enables efficient learning in high-dimensional settings

Adapts to distribution shifts effectively

Abstract

Libraries for supervised classification have enabled the wide-spread usage of machine learning methods. Existing libraries, such as scikit-learn, caret, and mlpack, implement techniques based on the classical empirical risk minimization (ERM) approach. We present a Python library, MRCpy, that implements minimax risk classifiers (MRCs) based on the robust risk minimization (RRM) approach. The library offers multiple variants of MRCs that can provide performance guarantees, enable efficient learning in high dimensions, and adapt to distribution shifts. MRCpy follows an object-oriented approach and adheres to the standards of popular Python libraries, such as scikit-learn, facilitating readability and easy usage together with a seamless integration with other libraries. The source code is available under the GPL-3.0 license at https://github.com/MachineLearningBCAM/MRCpy.