Alleviating label switching with optimal transport
© 2019 Neural information processing systems foundation. All rights reserved. Label switching is a phenomenon arising in mixture model posterior inference that prevents one from meaningfully assessing posterior statistics using standard Monte Carlo procedures. This issue arises due to invariance of...
| Príomhchruthaitheoirí: | , , , , , |
|---|---|
| Rannpháirtithe: | |
| Formáid: | Alt |
| Teanga: | English |
| Foilsithe / Cruthaithe: |
2022
|
| Rochtain ar líne: | https://hdl.handle.net/1721.1/137353.2 |
| _version_ | 1826191446924853248 |
|---|---|
| author | Monteiller, Pierre Claici, Sebastian Chien, Edward Mirzazadeh, Farzaneh Solomon, Justin Yurochkin, Mikhail |
| author2 | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory |
| author_facet | Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory Monteiller, Pierre Claici, Sebastian Chien, Edward Mirzazadeh, Farzaneh Solomon, Justin Yurochkin, Mikhail |
| author_sort | Monteiller, Pierre |
| collection | MIT |
| description | © 2019 Neural information processing systems foundation. All rights reserved. Label switching is a phenomenon arising in mixture model posterior inference that prevents one from meaningfully assessing posterior statistics using standard Monte Carlo procedures. This issue arises due to invariance of the posterior under actions of a group; for example, permuting the ordering of mixture components has no effect on the likelihood. We propose a resolution to label switching that leverages machinery from optimal transport. Our algorithm efficiently computes posterior statistics in the quotient space of the symmetry group. We give conditions under which there is a meaningful solution to label switching and demonstrate advantages over alternative approaches on simulated and real data. |
| first_indexed | 2024-09-23T08:56:17Z |
| format | Article |
| id | mit-1721.1/137353.2 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T08:56:17Z |
| publishDate | 2022 |
| record_format | dspace |
| spelling | mit-1721.1/137353.22022-01-03T16:40:51Z Alleviating label switching with optimal transport Monteiller, Pierre Claici, Sebastian Chien, Edward Mirzazadeh, Farzaneh Solomon, Justin Yurochkin, Mikhail Massachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratory MIT-IBM Watson AI Lab © 2019 Neural information processing systems foundation. All rights reserved. Label switching is a phenomenon arising in mixture model posterior inference that prevents one from meaningfully assessing posterior statistics using standard Monte Carlo procedures. This issue arises due to invariance of the posterior under actions of a group; for example, permuting the ordering of mixture components has no effect on the likelihood. We propose a resolution to label switching that leverages machinery from optimal transport. Our algorithm efficiently computes posterior statistics in the quotient space of the symmetry group. We give conditions under which there is a meaningful solution to label switching and demonstrate advantages over alternative approaches on simulated and real data. Army Research Office (Grant W911NF1710068) Air Force Office of Scientific Research (Award FA9550-19-1-031) National Science Foundation (Grant IIS-1838071) 2022-01-03T16:40:50Z 2021-11-04T15:54:19Z 2022-01-03T16:40:50Z 2019-12 2021-03-26T14:22:51Z Article http://purl.org/eprint/type/ConferencePaper https://hdl.handle.net/1721.1/137353.2 2019. "Alleviating label switching with optimal transport." Advances in Neural Information Processing Systems, 32. en https://papers.nips.cc/paper/2019/hash/c2ae5cb2426d96ed19a50b0b7d7c8e11-Abstract.html Advances in Neural Information Processing Systems Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. application/octet-stream Neural Information Processing Systems (NIPS) |
| spellingShingle | Monteiller, Pierre Claici, Sebastian Chien, Edward Mirzazadeh, Farzaneh Solomon, Justin Yurochkin, Mikhail Alleviating label switching with optimal transport |
| title | Alleviating label switching with optimal transport |
| title_full | Alleviating label switching with optimal transport |
| title_fullStr | Alleviating label switching with optimal transport |
| title_full_unstemmed | Alleviating label switching with optimal transport |
| title_short | Alleviating label switching with optimal transport |
| title_sort | alleviating label switching with optimal transport |
| url | https://hdl.handle.net/1721.1/137353.2 |
| work_keys_str_mv | AT monteillerpierre alleviatinglabelswitchingwithoptimaltransport AT claicisebastian alleviatinglabelswitchingwithoptimaltransport AT chienedward alleviatinglabelswitchingwithoptimaltransport AT mirzazadehfarzaneh alleviatinglabelswitchingwithoptimaltransport AT solomonjustin alleviatinglabelswitchingwithoptimaltransport AT yurochkinmikhail alleviatinglabelswitchingwithoptimaltransport |