Yes, but did it work?: Evaluating variational inference
| dc.contributor | Aalto-yliopisto | fi |
| dc.contributor | Aalto University | en |
| dc.contributor.author | Yao, Yuling | en_US |
| dc.contributor.author | Vehtari, Aki | en_US |
| dc.contributor.author | Simpson, Daniel | en_US |
| dc.contributor.author | Gelman, Andrew | en_US |
| dc.contributor.department | Department of Computer Science | en |
| dc.contributor.editor | Dy, Jennifer | en_US |
| dc.contributor.editor | Krause, Andreas | en_US |
| dc.contributor.groupauthor | Probabilistic Machine Learning | en |
| dc.contributor.groupauthor | Helsinki Institute for Information Technology (HIIT) | en |
| dc.contributor.groupauthor | Professorship Vehtari Aki | en |
| dc.contributor.organization | University of Toronto | en_US |
| dc.contributor.organization | Columbia University | en_US |
| dc.date.accessioned | 2019-07-30T07:20:18Z | |
| dc.date.available | 2019-07-30T07:20:18Z | |
| dc.date.issued | 2018-01-01 | en_US |
| dc.description.abstract | While it's always possible to compute a variational approximation to a posterior distribution, it can be difficult to discover problems with this approximation". We propose two diagnostic algorithms to alleviate this problem. The Paretosmoothed importance sampling (PSIS) diagnostic gives a goodness of fit measurement for joint distributions, while simultaneously improving the error in the estimate. The variational simulationbased calibration (VSBC) assesses the average performance of point estimates. | en |
| dc.description.version | Peer reviewed | en |
| dc.format.extent | 9 | |
| dc.format.mimetype | application/pdf | en_US |
| dc.identifier.citation | Yao, Y, Vehtari, A, Simpson, D & Gelman, A 2018, Yes, but did it work? Evaluating variational inference. in J Dy & A Krause (eds), 35th International Conference on Machine Learning, ICML 2018. vol. 12, Proceedings of Machine Learning Research, no. 80, International Machine Learning Society, pp. 8887-8895, International Conference on Machine Learning, Stockholm, Sweden, 10/07/2018. < http://proceedings.mlr.press/v80/yao18a.html > | en |
| dc.identifier.isbn | 9781510867963 | |
| dc.identifier.issn | 1938-7228 | |
| dc.identifier.other | PURE UUID: cda14f54-6e81-4974-af57-e1a87ed1f532 | en_US |
| dc.identifier.other | PURE ITEMURL: https://research.aalto.fi/en/publications/cda14f54-6e81-4974-af57-e1a87ed1f532 | en_US |
| dc.identifier.other | PURE LINK: http://www.scopus.com/inward/record.url?scp=85057318872&partnerID=8YFLogxK | |
| dc.identifier.other | PURE LINK: http://proceedings.mlr.press/v80/yao18a.html | en_US |
| dc.identifier.other | PURE FILEURL: https://research.aalto.fi/files/35131932/yao18a.pdf | en_US |
| dc.identifier.uri | https://aaltodoc.aalto.fi/handle/123456789/39504 | |
| dc.identifier.urn | URN:NBN:fi:aalto-201907304559 | |
| dc.language.iso | en | en |
| dc.relation.ispartof | International Conference on Machine Learning | en |
| dc.relation.ispartof | INTERNATIONAL CONFERENCE ON MACHINE LEARNING | fin |
| dc.relation.ispartofseries | 35th International Conference on Machine Learning, ICML 2018 | en |
| dc.relation.ispartofseries | Volume 12, pp. 8887-8895 | en |
| dc.relation.ispartofseries | Proceedings of Machine Learning Research ; 80 | en |
| dc.rights | openAccess | en |
| dc.title | Yes, but did it work?: Evaluating variational inference | en |
| dc.type | A4 Artikkeli konferenssijulkaisussa | fi |
| dc.type.version | publishedVersion |