NeurIPS 2019
Sun Dec 8th through Sat the 14th, 2019 at Vancouver Convention Center
Paper ID:3743
Title:Interior-Point Methods Strike Back: Solving the Wasserstein Barycenter Problem

The reviewers converged to a consensus of accepting this paper! Congratulations. In your revision, please address all comments in the reviews and include promised experiments mentioned in the rebuttal. One condition for us to accept your revision: Please add a clear derivation of the previous best run times for this problem. Also revise in light of the fact that the current method of computing the previous best run time (as they described in the rebuttal) is not quite correct, since it does not use the current fastest LP algorithms for the computations (naive methods have issues that were circumvented by the papers mentioned in the review).