On Linear Optimization over Wasserstein Balls
Preprint
Abstract
Wasserstein balls, which contain all probability measures within a pre-specified Wasserstein distance to a reference measure, have recently enjoyed wide popularity in the distributionally robust optimization and machine learning communities to formulate and solve data-driven optimization problems with rigorous statistical guarantees. In this technical note we prove that the Wasserstein ball is weakly compact under mild conditions, and we offer necessary and sufficient conditions for the existence of optimal solutions. We also characterize the sparsity of solutions if the Wasserstein ball is centred at a discrete reference measure. In comparison with the existing literature, which has proved similar results under different conditions, our proofs are self-contained and shorter, yet mathematically rigorous, and our necessary and sufficient conditions for the existence of optimal solutions are easily verifiable in practice.Keywords
All Related Versions
This publication has 6 references indexed in Scilit:
- Quantifying Distributional Model Risk via Optimal TransportMathematics of Operations Research, 2019
- Quantitative stability analysis for minimax distributionally robust risk optimizationMathematical Programming, 2018
- Data-driven risk-averse stochastic optimization with Wasserstein metricOperations Research Letters, 2018
- Extreme points of a ball about a measure with finite supportCommunications in Mathematical Sciences, 2017
- On the extreme points of moments setsMathematical Methods of Operations Research, 2016
- Wasserstein metric and subordinationStudia Mathematica, 2008