Optimal Transport for Gaussian Mixture Models

Authors:

Abstract
We introduce an optimal mass transport framework on the space of Gaussian mixture models. These models are widely used in statistical inference. Specifically, we treat the Gaussian mixture models as a submanifold of probability densities equipped with the Wasserstein metric. The topology induced by optimal transport is highly desirable and natural because, in contrast to total variation and other metrics, the Wasserstein metric is weakly continuous (i.e., convergence is equivalent to the convergence of moments). Thus, our approach provides natural ways to compare, interpolate, and average Gaussian mixture models. Moreover, the approach has low computational complexity. Different aspects of the framework are discussed, and examples are presented for illustration purposes.
Funding Information
  • National Science Foundation (1665031, 1807664, 1839441, 1901599)
  • Air Force Office of Scientific Research (FA9550-17-1-0435)
  • Army Research Office (W911NF-17-1-0429)
  • National Center for Research Resources (P41-RR-013218)
  • National Institute of Biomedical Imaging and Bioengineering (P41-EB-015902)
  • National Cancer Institute (1U24CA18092401A1)
  • NIA (R01 AG053991)
  • Breast Cancer Research Foundation

This publication has 23 references indexed in Scilit: