DRO: A Python Library for Distributionally Robust Optimization in Machine Learning

Jiashuo Liu, Tianyu Wang, Henry Lam, Hongseok Namkoong, Jose Blanchet

Keywords: distributionally robust optimization, computational software, distribution shift

TL;DR: We create an open-source Python library for various distributionally robust optimization for regression and classification problems.

Abstract:

We introduce **dro**, an open-source Python library for distributionally robust optimization (DRO) for regression and classification problems. The library implements 14 DRO formulations and 9 backbone models, enabling 79 distinct DRO methods. Furthermore, **dro** is compatible with both **scikit-learn** and **PyTorch**. Through vectorization and optimization approximation techniques, **dro** reduces runtime by 10x to over 1000x compared to baseline implementations on large-scale datasets.
Comprehensive documentation is available at https://python-dro.org.

Submission Number: 13