Wasserstein Distributionally Robust Optimization with Heterogeneous Data Sources

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Rychener, Yves, Esteban-Perez, Adrian, Morales, Juan M., Kuhn, Daniel
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914950429016064
author Rychener, Yves
Esteban-Perez, Adrian
Morales, Juan M.
Kuhn, Daniel
author_facet Rychener, Yves
Esteban-Perez, Adrian
Morales, Juan M.
Kuhn, Daniel
contents We study decision problems under uncertainty, where the decision-maker has access to $K$ data sources that carry {\em biased} information about the underlying risk factors. The biases are measured by the mismatch between the risk factor distribution and the $K$ data-generating distributions with respect to an optimal transport (OT) distance. In this situation the decision-maker can exploit the information contained in the biased samples by solving a distributionally robust optimization (DRO) problem, where the ambiguity set is defined as the intersection of $K$ OT neighborhoods, each of which is centered at the empirical distribution on the samples generated by a biased data source. We show that if the decision-maker has a prior belief about the biases, then the out-of-sample performance of the DRO solution can improve with $K$ -- irrespective of the magnitude of the biases. We also show that, under standard convexity assumptions, the proposed DRO problem is computationally tractable if either $K$ or the dimension of the risk factors is kept constant.
format Preprint
id arxiv_https___arxiv_org_abs_2407_13582
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Wasserstein Distributionally Robust Optimization with Heterogeneous Data Sources
Rychener, Yves
Esteban-Perez, Adrian
Morales, Juan M.
Kuhn, Daniel
Optimization and Control
Probability
Statistics Theory
We study decision problems under uncertainty, where the decision-maker has access to $K$ data sources that carry {\em biased} information about the underlying risk factors. The biases are measured by the mismatch between the risk factor distribution and the $K$ data-generating distributions with respect to an optimal transport (OT) distance. In this situation the decision-maker can exploit the information contained in the biased samples by solving a distributionally robust optimization (DRO) problem, where the ambiguity set is defined as the intersection of $K$ OT neighborhoods, each of which is centered at the empirical distribution on the samples generated by a biased data source. We show that if the decision-maker has a prior belief about the biases, then the out-of-sample performance of the DRO solution can improve with $K$ -- irrespective of the magnitude of the biases. We also show that, under standard convexity assumptions, the proposed DRO problem is computationally tractable if either $K$ or the dimension of the risk factors is kept constant.
title Wasserstein Distributionally Robust Optimization with Heterogeneous Data Sources
topic Optimization and Control
Probability
Statistics Theory
url https://arxiv.org/abs/2407.13582