AISTATS 2023poster11 citations

Bures-Wasserstein Barycenters and Low-Rank Matrix Recovery

Tyler Maunu, Thibaut Le Gouic, Philippe Rigollet

Abstract

We revisit the problem of recovering a low-rank positive semidefinite matrix from rank-one projections using tools from optimal transport. More specifically, we show that a variational formulation of this problem is equivalent to computing a Wasserstein barycenter. In turn, this new perspective enables the development of new geometric first-order methods with strong convergence guarantees in Bures-Wasserstein distance. Experiments on simulated data demonstrate the advantages of our new methodology over existing methods.

BibTeX
@InProceedings{pmlr-v206-maunu23a,
  title = 	 {Bures-Wasserstein Barycenters and Low-Rank Matrix Recovery},
  author =       {Maunu, Tyler and Le Gouic, Thibaut and Rigollet, Philippe},
  booktitle = 	 {Proceedings of The 26th International Conference on Artificial Intelligence and Statistics},
  pages = 	 {8183--8210},
  year = 	 {2023},
  editor = 	 {Ruiz, Francisco and Dy, Jennifer and van de Meent, Jan-Willem},
  volume = 	 {206},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {25--27 Apr},
  publisher =    {PMLR},
  pdf = 	 {https://proceedings.mlr.press/v206/maunu23a/maunu23a.pdf},
  url = 	 {https://proceedings.mlr.press/v206/maunu23a.html},
  abstract = 	 {We revisit the problem of recovering a low-rank positive semidefinite matrix from rank-one projections using tools from optimal transport. More specifically, we show that a variational formulation of this problem is equivalent to computing a Wasserstein barycenter. In turn, this new perspective enables the development of new geometric first-order methods with strong convergence guarantees in Bures-Wasserstein distance. Experiments on simulated data demonstrate the advantages of our new methodology over existing methods.}
}