Unifying Distributionally Robust Optimization via Optimal Transport Theory

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Blanchet, Jose, Kuhn, Daniel, Li, Jiajin, Taskesen, Bahar
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909969879662592
author Blanchet, Jose
Kuhn, Daniel
Li, Jiajin
Taskesen, Bahar
author_facet Blanchet, Jose
Kuhn, Daniel
Li, Jiajin
Taskesen, Bahar
contents In recent years, two prominent paradigms have shaped distributionally robust optimization (DRO), modeling distributional ambiguity through $ϕ$-divergences and Wasserstein distances, respectively. While the former focuses on ambiguity in likelihood ratios, the latter emphasizes ambiguity in outcomes and uses a transportation cost function to capture geometric structure in the outcome space. This paper proposes a unified framework that bridges these approaches by leveraging optimal transport (OT) with conditional moment constraints. Our formulation enables adversarial distributions to jointly perturb likelihood ratios and outcomes, yielding a generalized OT coupling between the nominal and perturbed distributions. We further establish key duality results and develop tractable reformulations that highlight the practical power of our unified approach.
format Preprint
id arxiv_https___arxiv_org_abs_2308_05414
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Unifying Distributionally Robust Optimization via Optimal Transport Theory
Blanchet, Jose
Kuhn, Daniel
Li, Jiajin
Taskesen, Bahar
Optimization and Control
Machine Learning
In recent years, two prominent paradigms have shaped distributionally robust optimization (DRO), modeling distributional ambiguity through $ϕ$-divergences and Wasserstein distances, respectively. While the former focuses on ambiguity in likelihood ratios, the latter emphasizes ambiguity in outcomes and uses a transportation cost function to capture geometric structure in the outcome space. This paper proposes a unified framework that bridges these approaches by leveraging optimal transport (OT) with conditional moment constraints. Our formulation enables adversarial distributions to jointly perturb likelihood ratios and outcomes, yielding a generalized OT coupling between the nominal and perturbed distributions. We further establish key duality results and develop tractable reformulations that highlight the practical power of our unified approach.
title Unifying Distributionally Robust Optimization via Optimal Transport Theory
topic Optimization and Control
Machine Learning
url https://arxiv.org/abs/2308.05414