swMATH ID: 39713
Software Authors: Kartheek Bondugula, Santiago Mazuelas, Aritz Pérez
Description: MRCpy: A Library for Minimax Risk Classifiers. Existing libraries for supervised classification implement techniques that are based on empirical risk minimization and utilize surrogate losses. We present MRCpy library that implements minimax risk classifiers (MRCs) that are based on robust risk minimization and can utilize 0-1-loss. Such techniques give rise to a manifold of classification methods that can provide tight bounds on the expected loss. MRCpy provides a unified interface for different variants of MRCs and follows the standards of popular Python libraries. The presented library also provides implementation for popular techniques that can be seen as MRCs such as L1-regularized logistic regression, zero-one adversarial, and maximum entropy machines. In addition, MRCpy implements recent feature mappings such as Fourier, ReLU, and threshold features. The library is designed with an object-oriented approach that facilitates collaborators and users.
Homepage: https://machinelearningbcam.github.io/MRCpy/
Source Code: https://github.com/MachineLearningBCAM/MRCpy
Dependencies: Python
Keywords: Machine Learning; arXiv_stat.ML; arXiv_cs.LG; MRCpy; Minimax Risk Classifiers; Python; supervised classification; robust risk minimization; feature mappings
Related Software: flake8; SciPy; NumPy; CVXPY; Scikit; Python
Cited in: 0 Publications

Standard Articles

1 Publication describing the Software Year
MRCpy: A Library for Minimax Risk Classifiers
Kartheek Bondugula, Santiago Mazuelas, Aritz Pérez