Efficient Approximation Algorithm for Computing Wasserstein Barycenter under Euclidean Metric
Pankaj K. Agarwal, Sharath Raghvendra, Pouyan Shirzadian, Keegan Yao · Society for Industrial and Applied Mathematics eBooks · 2025
Given a set of probability distributions, the Wasserstein barycenter problem asks to compute a distribution that minimizes the average Wasserstein distance, or optimal transport cost, from all the input distributions. Wasserstein barycenters preserve common geometric features of the input distributions, making them useful in machine learning and data analytics tasks.