The GenCol algorithm for high-dimensional optimal transport: general formulation and application to barycenters and Wasserstein splines

This work provides a computational tool for researchers in applied mathematics and machine learning dealing with optimal transport, though it appears incremental as an extension of an existing algorithm.

The authors tackled the problem of high-dimensional multi-marginal optimal transport by extending the GenCol algorithm from symmetric to general formulations, resulting in accurate mesh-free calculations of Wasserstein barycenters and cubic splines.