Show simple item record

dc.contributor.authorTrosten, Daniel Johansen
dc.contributor.authorLøkse, Sigurd Eivindson
dc.contributor.authorJenssen, Robert
dc.contributor.authorKampffmeyer, Michael
dc.date.accessioned2022-03-10T12:44:22Z
dc.date.available2022-03-10T12:44:22Z
dc.date.issued2021-11-13
dc.description.abstractAligning distributions of view representations is a core component of today’s state of the art models for deep multi-view clustering. However, we identify several drawbacks with naïvely aligning representation distributions. We demonstrate that these drawbacks both lead to less separable clusters in the representation space, and inhibit the model’s ability to prioritize views. Based on these observations, we develop a simple baseline model for deep multi-view clustering. Our baseline model avoids representation alignment altogether, while performing similar to, or better than, the current state of the art. We also expand our baseline model by adding a contrastive learning component. This introduces a selective alignment procedure that preserves the model’s ability to prioritize views. Our experiments show that the contrastive learning component enhances the baseline model, improving on the current state of the art by a large margin on several datasets.en_US
dc.identifier.citationTrosten DJ, Løkse S, Jenssen R, Kampffmeyer MC. Reconsidering Representation Alignment for Multi-View Clustering. Computer Vision and Pattern Recognition. 2021en_US
dc.identifier.cristinIDFRIDAID 1938162
dc.identifier.doi10.1109/CVPR46437.2021.00131
dc.identifier.issn1063-6919
dc.identifier.urihttps://hdl.handle.net/10037/24371
dc.language.isoengen_US
dc.publisherIEEEen_US
dc.relation.ispartofTrosten, D.J. (2023). Improving Representation Learning for Deep Clustering and Few-shot Learning. (Doctoral thesis). <a href=https://hdl.handle.net/10037/29847>https://hdl.handle.net/10037/29847</a>.
dc.relation.journalComputer Vision and Pattern Recognition
dc.relation.projectIDNorges forskningsråd: 309439en_US
dc.relation.projectIDNorges forskningsråd: 315029en_US
dc.relation.projectIDNorges forskningsråd: 303514en_US
dc.rights.accessRightsopenAccessen_US
dc.rights.holderCopyright 2021 The Author(s)en_US
dc.titleReconsidering Representation Alignment for Multi-View Clusteringen_US
dc.type.versionacceptedVersionen_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