Sinkhorn Distributionally Robust Optimization
Optimization and Control
2025-03-27 v5 Machine Learning
Machine Learning
Abstract
We study distributionally robust optimization with Sinkhorn distance -- a variant of Wasserstein distance based on entropic regularization. We derive a convex programming dual reformulation for general nominal distributions, transport costs, and loss functions. To solve the dual reformulation, we develop a stochastic mirror descent algorithm with biased subgradient estimators and derive its computational complexity guarantees. Finally, we provide numerical examples using synthetic and real data to demonstrate its superior performance.
Cite
@article{arxiv.2109.11926,
title = {Sinkhorn Distributionally Robust Optimization},
author = {Jie Wang and Rui Gao and Yao Xie},
journal= {arXiv preprint arXiv:2109.11926},
year = {2025}
}
Comments
55 pages, 15 figures