Show simple item record

dc.contributor.authorGautam, Srishti
dc.contributor.authorBoubekki, Ahcene
dc.contributor.authorHansen, Stine
dc.contributor.authorSalahuddin, Suaiba Amina
dc.contributor.authorJenssen, Robert
dc.contributor.authorHohne, Marina Marie-Claire
dc.contributor.authorKampffmeyer, Michael
dc.date.accessioned2024-03-14T11:22:17Z
dc.date.available2024-03-14T11:22:17Z
dc.date.issued2022-10-15
dc.description.abstractThe need for interpretable models has fostered the development of self-explainable classifiers. Prior approaches are either based on multi-stage optimization schemes, impacting the predictive performance of the model, or produce explanations that are not transparent, trustworthy or do not capture the diversity of the data. To address these shortcomings, we propose ProtoVAE, a variational autoencoder-based framework that learns class-specific prototypes in an end-to-end manner and enforces trustworthiness and diversity by regularizing the representation space and introducing an orthonormality constraint. Finally, the model is designed to be transparent by directly incorporating the prototypes into the decision process. Extensive comparisons with previous self-explainable approaches demonstrate the superiority of ProtoVAE, highlighting its ability to generate trustworthy and diverse explanations, while not degrading predictive performance.en_US
dc.descriptionSource at <a href=https://nips.cc/>https://nips.cc/</a>.en_US
dc.identifier.citationGautam S, Boubekki A, Hansen S, Salahuddin SA, Jenssen R, Hohne MM, Kampffmeyer MC. ProtoVAE: A Trustworthy Self-Explainable Prototypical Variational Model. Advances in Neural Information Processing Systems. 2022en_US
dc.identifier.cristinIDFRIDAID 2084871
dc.identifier.issn1049-5258
dc.identifier.urihttps://hdl.handle.net/10037/33160
dc.language.isoengen_US
dc.relation.journalAdvances in Neural Information Processing Systems
dc.relation.projectIDNorges forskningsråd: 315029en_US
dc.relation.projectIDNorges forskningsråd: 309439en_US
dc.relation.projectIDNorges forskningsråd: 303514en_US
dc.rights.accessRightsopenAccessen_US
dc.rights.holderCopyright 2022 The Author(s)en_US
dc.titleProtoVAE: A Trustworthy Self-Explainable Prototypical Variational Modelen_US
dc.type.versionpublishedVersionen_US
dc.typeJournal articleen_US
dc.typeTidsskriftartikkelen_US
dc.typePeer revieweden_US


File(s) in this item

Thumbnail

This item appears in the following collection(s)

Show simple item record