Stochastic wasserstein barycenters

© 2018 35th International Conference on Machine Learning, ICML 2018. All rights reserved. Wi present a stochastic algorithm to compute the baryccntcr of a set of probability distributions under the Wasscrstcin metric from optimal transport Unlike previous approaches,our method extends to continuous...

Full description

Bibliographic Details
Main Authors: Solomon, Justin, Chien, Edward, Claici, Sebastian
Other Authors: Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Format: Article
Language:English
Published: 2021
Online Access:https://hdl.handle.net/1721.1/137895.2
_version_ 1811092333705297920
author Solomon, Justin
Chien, Edward
Claici, Sebastian
author2 Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
author_facet Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory
Solomon, Justin
Chien, Edward
Claici, Sebastian
author_sort Solomon, Justin
collection MIT
description © 2018 35th International Conference on Machine Learning, ICML 2018. All rights reserved. Wi present a stochastic algorithm to compute the baryccntcr of a set of probability distributions under the Wasscrstcin metric from optimal transport Unlike previous approaches,our method extends to continuous input distributions and allows the support of the baryccntcr to be adjusted in each iteration. VVc tacklc the problem without rcgu- larization, allowing us to rccovcr a much sharper output; We give examples where our algorithm recovers a more meaningful baryccntcr than previous work. Our method is versatile and can be extended to applications such as generating super samples from a given distribution and recovering blue noise approximations.
first_indexed 2024-09-23T15:16:13Z
format Article
id mit-1721.1/137895.2
institution Massachusetts Institute of Technology
language English
last_indexed 2024-09-23T15:16:13Z
publishDate 2021
record_format dspace
spelling mit-1721.1/137895.22021-11-09T18:35:58Z Stochastic wasserstein barycenters Solomon, Justin Chien, Edward Claici, Sebastian Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory © 2018 35th International Conference on Machine Learning, ICML 2018. All rights reserved. Wi present a stochastic algorithm to compute the baryccntcr of a set of probability distributions under the Wasscrstcin metric from optimal transport Unlike previous approaches,our method extends to continuous input distributions and allows the support of the baryccntcr to be adjusted in each iteration. VVc tacklc the problem without rcgu- larization, allowing us to rccovcr a much sharper output; We give examples where our algorithm recovers a more meaningful baryccntcr than previous work. Our method is versatile and can be extended to applications such as generating super samples from a given distribution and recovering blue noise approximations. Army Research Office (Grant W911NF-12- R0011) 2021-11-09T18:35:57Z 2021-11-09T14:55:47Z 2021-11-09T18:35:57Z 2018 2019-07-10T12:24:25Z Article http://purl.org/eprint/type/ConferencePaper https://hdl.handle.net/1721.1/137895.2 Solomon, Justin, Chien, Edward and Claici, Sebastian. 2018. "Stochastic wasserstein barycenters." en http://proceedings.mlr.press/v80/claici18a/claici18a.pdf Creative Commons Attribution-Noncommercial-Share Alike http://creativecommons.org/licenses/by-nc-sa/4.0/ application/octet-stream arXiv
spellingShingle Solomon, Justin
Chien, Edward
Claici, Sebastian
Stochastic wasserstein barycenters
title Stochastic wasserstein barycenters
title_full Stochastic wasserstein barycenters
title_fullStr Stochastic wasserstein barycenters
title_full_unstemmed Stochastic wasserstein barycenters
title_short Stochastic wasserstein barycenters
title_sort stochastic wasserstein barycenters
url https://hdl.handle.net/1721.1/137895.2
work_keys_str_mv AT solomonjustin stochasticwassersteinbarycenters
AT chienedward stochasticwassersteinbarycenters
AT claicisebastian stochasticwassersteinbarycenters