# Decomposition Algorithm for Distributionally Robust Optimization using Wasserstein Metric

arXiv: https://arxiv.org/abs/1704.03920 · HTML: https://mlsanity.com/paper/1704.03920/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 12 Apr 2017.

## Built on by (most cited first, 4 of 4)

- [Distributionally Robust Learning](https://mlsanity.com/paper/2108.08993/) (2021)
- [On the Benefits of Over-parameterization for Out-of-Distribution Generalization](https://mlsanity.com/paper/2403.17592/) (2024)
- [Distributionally Robust Inverse Reinforcement Learning for Identifying Multi-Agent Coordinated Sensing](https://mlsanity.com/paper/2409.14542/) (2024)
- [Multi-Agent Inverse Reinforcement Learning for Identifying Pareto-Efficient Coordination -- A Distributionally Robust Approach](https://mlsanity.com/paper/2509.08956/) (2025)

## Abstract (arXiv)

We study distributionally robust optimization (DRO) problems where the ambiguity set is defined using the Wasserstein metric. We show that this class of DRO problems can be reformulated as semi-infinite programs. We give an exchange method to solve the reformulated problem for the general nonlinear model, and a central cutting-surface method for the convex case, assuming that we have a separation oracle. We used a distributionally robust generalization of the logistic regression model to test our algorithm. Numerical experiments on the distributionally robust logistic regression models show that the number of oracle calls are typically 20 ? 50 to achieve 5-digit precision. The solution found by the model is generally better in its ability to predict with a smaller standard error.