Show simple item record

dc.contributor.authorHazan, Tamir
dc.contributor.authorMaji, Subhransu
dc.contributor.authorKeshet, Joseph
dc.contributor.authorJaakkola, Tommi S.
dc.date.accessioned2015-12-17T00:46:23Z
dc.date.available2015-12-17T00:46:23Z
dc.date.issued2013
dc.identifier.issn1049-5258
dc.identifier.urihttp://hdl.handle.net/1721.1/100402
dc.description.abstractIn this work we develop efficient methods for learning random MAP predictors for structured label problems. In particular, we construct posterior distributions over perturbations that can be adjusted via stochastic gradient methods. We show that every smooth posterior distribution would suffice to define a smooth PAC-Bayesian risk bound suitable for gradient methods. In addition, we relate the posterior distributions to computational properties of the MAP predictors. We suggest multiplicative posteriors to learn super-modular potential functions that accompany specialized MAP predictors such as graph-cuts. We also describe label-augmented posterior models that can use efficient MAP approximations, such as those arising from linear program relaxations.en_US
dc.language.isoen_US
dc.publisherNeural Information Processing Systemsen_US
dc.relation.isversionofhttp://papers.nips.cc/paper/5066-learning-efficient-random-maximum-a-posteriori-predictors-with-non-decomposable-loss-functionsen_US
dc.rightsArticle 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.en_US
dc.sourceMIT web domainen_US
dc.titleLearning efficient random maximum a-posteriori predictors with non-decomposable loss functionsen_US
dc.typeArticleen_US
dc.identifier.citationHazan, Tamir, Subhransu Maji, Joseph Keshet, and Tommi Jaakkola. "Learning efficient random maximum a-posteriori predictors with non-decomposable loss functions." Advances in Neural Information Processing Systems (NIPS 2013).en_US
dc.contributor.departmentMassachusetts Institute of Technology. Computer Science and Artificial Intelligence Laboratoryen_US
dc.contributor.departmentMassachusetts Institute of Technology. Department of Electrical Engineering and Computer Scienceen_US
dc.contributor.mitauthorJaakkola, Tommi S.en_US
dc.relation.journalAdvances in Neural Information Processing Systems (NIPS)en_US
dc.eprint.versionFinal published versionen_US
dc.type.urihttp://purl.org/eprint/type/ConferencePaperen_US
eprint.statushttp://purl.org/eprint/status/NonPeerRevieweden_US
dspace.orderedauthorsHazan, Tamir; Maji, Subhransu; Keshet, Joseph; Jaakkola, Tommien_US
dc.identifier.orcidhttps://orcid.org/0000-0002-2199-0379
mit.licensePUBLISHER_POLICYen_US


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record