Skip to content

Questions about support sets for Wasserstein barycenters #367

Answered by rflamary
sandeshAdhikary asked this question in Q&A
Discussion options

You must be logged in to vote
  1. There is for the moment no "general ground metrics" free_support_barycenter because the algorithm implemented is a block coordinate descent that requires a closed form solution for the update of the position which is not available for any losses. For general loss you can indeed implement a gradient descent form a sum of losses returned by ot.emd2 when using differentiable loss matrices.

  2. The ot.barycenterfunction supposes that the support for all the distributions in the barycenter is the same (because a ground cost matrix is necessary and the j index in M_i,j index the positions in the distributions) but the barycenter itself does not need to be the same support as the distributions (…

Replies: 1 comment 1 reply

Comment options

You must be logged in to vote
1 reply
@sandeshAdhikary
Comment options

Answer selected by sandeshAdhikary
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Category
Q&A
Labels
None yet
2 participants