M
Marco Cuturi
Researcher at Google
Publications - 155
Citations - 12954
Marco Cuturi is an academic researcher from Google. The author has contributed to research in topics: Computer science & Metric (mathematics). The author has an hindex of 42, co-authored 141 publications receiving 9403 citations. Previous affiliations of Marco Cuturi include École Normale Supérieure & Mines ParisTech.
Papers
More filters
Proceedings Article
Regularized Optimal Transport is Ground Cost Adversarial
TL;DR: This work proposes a new interpretation of regularization as a robust mechanism, and shows using Fenchel duality that any convex regularization of OT can be interpreted as ground cost adversarial, which incidentally gives access to a robust dissimilarity measure on the ground space, which can in turn be used in other applications.
Journal ArticleDOI
Semidual Regularized Optimal Transport
Marco Cuturi,Gabriel Peyré +1 more
TL;DR: Variational problems that involve Wasserstein distances and more generally optimal transport theory are playing an increasingly important role in data sciences as discussed by the authors, and such problems can be used to solve data problems.
Proceedings Article
Fixed-Support Wasserstein Barycenters: Computational Hardness and Fast Algorithm
TL;DR: In this paper, the fixed-support Wasserstein barycenter problem (FS-WBP) was studied in the standard linear programming (LP) form, and a provably fast variant of the celebrated iterative Bregman projection (IBP) algorithm, named \textsc{FastIBP}, was developed, with a complexity bound of
Journal ArticleDOI
Multi-subject MEG/EEG source imaging with sparse multi-task regression
Hicham Janati,Hicham Janati,Thomas Bazeille,Thomas Bazeille,Bertrand Thirion,Bertrand Thirion,Marco Cuturi,Marco Cuturi,Alexandre Gramfort,Alexandre Gramfort +9 more
TL;DR: In this paper, the authors propose the Minimum Wasserstein Estimates (MWE) method, which is a joint regression method based on optimal transport (OT) metrics to promote spatial proximity on the cortical mantle.
Proceedings Article
Mapping kernels for trees
TL;DR: It is argued that most existing tree kernels, as well as many more that are presented for the first time in this paper, fall into a typology of kernels whose seemingly intricate computation can be efficiently factorized to yield polynomial time algorithms.