Show simple item record

hal.structure.identifierCEntre de REcherches en MAthématiques de la DEcision [CEREMADE]
dc.contributor.authorPedregosa, Fabian
dc.date.accessioned2018-02-19T12:39:22Z
dc.date.available2018-02-19T12:39:22Z
dc.date.issued2016
dc.identifier.urihttps://basepub.dauphine.fr/handle/123456789/17419
dc.language.isoenen
dc.subjectHyperparameter optimizationen
dc.subjectgradienten
dc.subject.ddc621.3en
dc.titleHyperparameter optimization with approximate gradienten
dc.typeCommunication / Conférence
dc.description.abstractenMost models in machine learning contain at least one hyperparameter to control for model complexity. Choosing an appropriate set of hyperparameters is both crucial in terms of model accuracy and computationally challenging. In this work we propose an algorithm for the optimization of continuous hyperparameters using inexact gradient information. An advantage of this method is that hyperparameters can be updated before model parameters have fully converged. We also give sufficient conditions for the global convergence of this method, based on regularity conditions of the involved functions and summability of errors. Finally, we validate the empirical performance of this method on the estimation of regularization constants of L2-regularized logistic regression and kernel Ridge regression. Empirical benchmarks indicate that our approach is highly competitive with respect to state of the art methods.en
dc.identifier.citationpages15en
dc.relation.ispartoftitleProceedings of the 33rd International Conference on Machine Learning, volume 48en
dc.relation.ispartofpublnameProceedings of Machine Learning Researchen
dc.relation.ispartofdate2016
dc.identifier.urlsitehttp://proceedings.mlr.press/v48/en
dc.subject.ddclabelTraitement du signalen
dc.relation.conftitleInternational Conference on Machine Learningen
dc.relation.confdate2016-06
dc.relation.confcityNew Yorken
dc.relation.confcountryUnited Statesen
dc.relation.forthcomingnonen
dc.description.ssrncandidatenonen
dc.description.halcandidatenonen
dc.description.readershiprechercheen
dc.description.audienceInternationalen
dc.relation.Isversionofjnlpeerreviewednonen
dc.relation.Isversionofjnlpeerreviewednonen
hal.author.functionaut


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record