Show simple item record

dc.contributor.authorMalinin, Andrey
dc.contributor.authorGales, Mark
dc.date.accessioned2019-01-12T00:32:13Z
dc.date.available2019-01-12T00:32:13Z
dc.date.issued2018-12-31
dc.identifier.issn1049-5258
dc.identifier.urihttps://www.repository.cam.ac.uk/handle/1810/287924
dc.description.abstractEstimating how uncertain an AI system is in its predictions is important to improve the safety of such systems. Uncertainty in predictive can result from uncertainty in model parameters, irreducible data uncertainty and uncertainty due to distributional mismatch between the test and training data distributions. Different actions might be taken depending on the source of the uncertainty so it is important to be able to distinguish between them. Recently, baseline tasks and metrics have been defined and several practical methods to estimate uncertainty developed. These methods, however, attempt to model uncertainty due to distributional mismatch either implicitly through model uncertainty or as data uncertainty. This work proposes a new framework for modeling predictive uncertainty called Prior Networks (PNs) which explicitly models distributional uncertainty. PNs do this by parameterizing a prior distribution over predictive distributions. This work focuses on uncertainty for classification and evaluates PNs on the tasks of identifying out-of-distribution (OOD) samples and detecting misclassification on the MNIST dataset, where they are found to outperform previous methods. Experiments on synthetic and MNIST data show that unlike previous non-Bayesian methods PNs are able to distinguish between data and distributional uncertainty.
dc.publisherCurran Associates, Inc.
dc.titlePredictive Uncertainty Estimation via Prior Networks
dc.typeConference Object
prism.endingPage7058
prism.publicationDate2018
prism.publicationNameNIPS'18: Proceedings of the 32nd International Conference on Neural Information Processing Systems
prism.startingPage7047
prism.volume31
dc.identifier.doi10.17863/CAM.35237
dcterms.dateAccepted2018-09-04
rioxxterms.versionofrecord10.17863/CAM.35237
rioxxterms.versionAM
rioxxterms.licenseref.urihttp://www.rioxx.net/licenses/all-rights-reserved
rioxxterms.licenseref.startdate2018-12-31
dc.contributor.orcidGales, Mark [0000-0002-5311-8219]
dc.publisher.urlhttps://papers.nips.cc/paper/7936-predictive-uncertainty-estimation-via-prior-networks
rioxxterms.typeConference Paper/Proceeding/Abstract
dc.identifier.urlhttps://papers.nips.cc/paper/7936-predictive-uncertainty-estimation-via-prior-networks
pubs.conference-nameNIPS 2018
pubs.conference-start-date2018-12-03
pubs.conference-finish-date2018-12-08
rioxxterms.freetoread.startdate2019-12-08


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record