$\texttt{skwdro}$: a library for Wasserstein distributionally robust machine learning

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Vincent, Florian, Azizian, Waïss, Iutzeler, Franck, Malick, Jérôme
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909986092744704
author Vincent, Florian
Azizian, Waïss
Iutzeler, Franck
Malick, Jérôme
author_facet Vincent, Florian
Azizian, Waïss
Iutzeler, Franck
Malick, Jérôme
contents We present skwdro, a Python library for training robust machine learning models. The library is based on distributionally robust optimization using Wasserstein distances, popular in optimal transport and machine learnings. The goal of the library is to make the training of robust models easier for a wide audience by proposing a wrapper for PyTorch modules, enabling model loss' robustification with minimal code changes. It comes along with scikit-learn compatible estimators for some popular objectives. The core of the implementation relies on an entropic smoothing of the original robust objective, in order to ensure maximal model flexibility. The library is available at https://github.com/iutzeler/skwdro and the documentation at https://skwdro.readthedocs.io.
format Preprint
id arxiv_https___arxiv_org_abs_2410_21231
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle $\texttt{skwdro}$: a library for Wasserstein distributionally robust machine learning
Vincent, Florian
Azizian, Waïss
Iutzeler, Franck
Malick, Jérôme
Machine Learning
Mathematical Software
Optimization and Control
90C17, 90C15
I.2.6; I.2.5; G.4; G.1.6
We present skwdro, a Python library for training robust machine learning models. The library is based on distributionally robust optimization using Wasserstein distances, popular in optimal transport and machine learnings. The goal of the library is to make the training of robust models easier for a wide audience by proposing a wrapper for PyTorch modules, enabling model loss' robustification with minimal code changes. It comes along with scikit-learn compatible estimators for some popular objectives. The core of the implementation relies on an entropic smoothing of the original robust objective, in order to ensure maximal model flexibility. The library is available at https://github.com/iutzeler/skwdro and the documentation at https://skwdro.readthedocs.io.
title $\texttt{skwdro}$: a library for Wasserstein distributionally robust machine learning
topic Machine Learning
Mathematical Software
Optimization and Control
90C17, 90C15
I.2.6; I.2.5; G.4; G.1.6
url https://arxiv.org/abs/2410.21231