Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386076172> ?p ?o ?g. }
- W4386076172 abstract "Clustering has been a major research topic in the field of machine learning, one to which Deep Learning has recently been applied with significant success. However, an aspect of clustering that is not addressed by existing deep clustering methods, is that of efficiently producing multiple, diverse partitionings for a given dataset. This is particularly important, as a diverse set of base clusterings are necessary for consensus clustering, which has been found to produce better and more robust results than relying on a single clustering. To address this gap, we propose Div-Clust, a diversity controlling loss that can be incorporated into existing deep clustering frameworks to produce multiple clusterings with the desired degree of diversity. We conduct experiments with multiple datasets and deep clustering frameworks and show that: a) our method effectively controls diversity across frameworks and datasets with very small additional computational cost, b) the sets of clusterings learned by DivClust include solutions that significantly outperform single-clustering baselines, and c) using an off-the-shelf consensus clustering algorithm, DivClust produces consensus clustering solutions that consistently outperform single-clustering baselines, effectively improving the performance of the base deep clustering framework. Code is available at https://github.com/ManiadisG/DivClust." @default.
- W4386076172 created "2023-08-23" @default.
- W4386076172 creator A5024224610 @default.
- W4386076172 creator A5031205865 @default.
- W4386076172 creator A5086692515 @default.
- W4386076172 date "2023-06-01" @default.
- W4386076172 modified "2023-09-27" @default.
- W4386076172 title "DivClust: Controlling Diversity in Deep Clustering" @default.
- W4386076172 cites W1907775068 @default.
- W4386076172 cites W2046025323 @default.
- W4386076172 cites W2107208924 @default.
- W4386076172 cites W2108343180 @default.
- W4386076172 cites W2111171370 @default.
- W4386076172 cites W2116984363 @default.
- W4386076172 cites W2121207090 @default.
- W4386076172 cites W2122491887 @default.
- W4386076172 cites W2132951961 @default.
- W4386076172 cites W2139280638 @default.
- W4386076172 cites W2144419338 @default.
- W4386076172 cites W2170718421 @default.
- W4386076172 cites W22892362 @default.
- W4386076172 cites W2293273145 @default.
- W4386076172 cites W2344987272 @default.
- W4386076172 cites W2529045002 @default.
- W4386076172 cites W2566414761 @default.
- W4386076172 cites W2585078009 @default.
- W4386076172 cites W2742098698 @default.
- W4386076172 cites W2762595523 @default.
- W4386076172 cites W2779692282 @default.
- W4386076172 cites W2788552663 @default.
- W4386076172 cites W2791146007 @default.
- W4386076172 cites W2809541107 @default.
- W4386076172 cites W2883725317 @default.
- W4386076172 cites W2907772339 @default.
- W4386076172 cites W2941460796 @default.
- W4386076172 cites W2952171869 @default.
- W4386076172 cites W2962852342 @default.
- W4386076172 cites W2965622195 @default.
- W4386076172 cites W2990500698 @default.
- W4386076172 cites W2994560339 @default.
- W4386076172 cites W2998345170 @default.
- W4386076172 cites W2998570830 @default.
- W4386076172 cites W3034363127 @default.
- W4386076172 cites W3034455162 @default.
- W4386076172 cites W3083270358 @default.
- W4386076172 cites W3087124270 @default.
- W4386076172 cites W3090103671 @default.
- W4386076172 cites W3106090541 @default.
- W4386076172 cites W3106709020 @default.
- W4386076172 cites W3108015556 @default.
- W4386076172 cites W3109374982 @default.
- W4386076172 cites W3110446398 @default.
- W4386076172 cites W3121769020 @default.
- W4386076172 cites W3128458717 @default.
- W4386076172 cites W3137513727 @default.
- W4386076172 cites W3157704640 @default.
- W4386076172 cites W3173944171 @default.
- W4386076172 cites W3183630498 @default.
- W4386076172 cites W3199244260 @default.
- W4386076172 cites W4214510096 @default.
- W4386076172 cites W4297970207 @default.
- W4386076172 cites W4312904550 @default.
- W4386076172 cites W4323345772 @default.
- W4386076172 doi "https://doi.org/10.1109/cvpr52729.2023.00333" @default.
- W4386076172 hasPublicationYear "2023" @default.
- W4386076172 type Work @default.
- W4386076172 citedByCount "0" @default.
- W4386076172 crossrefType "proceedings-article" @default.
- W4386076172 hasAuthorship W4386076172A5024224610 @default.
- W4386076172 hasAuthorship W4386076172A5031205865 @default.
- W4386076172 hasAuthorship W4386076172A5086692515 @default.
- W4386076172 hasConcept C119857082 @default.
- W4386076172 hasConcept C124101348 @default.
- W4386076172 hasConcept C154945302 @default.
- W4386076172 hasConcept C177264268 @default.
- W4386076172 hasConcept C186767784 @default.
- W4386076172 hasConcept C199360897 @default.
- W4386076172 hasConcept C202444582 @default.
- W4386076172 hasConcept C33704608 @default.
- W4386076172 hasConcept C33923547 @default.
- W4386076172 hasConcept C41008148 @default.
- W4386076172 hasConcept C73555534 @default.
- W4386076172 hasConcept C94641424 @default.
- W4386076172 hasConcept C9652623 @default.
- W4386076172 hasConceptScore W4386076172C119857082 @default.
- W4386076172 hasConceptScore W4386076172C124101348 @default.
- W4386076172 hasConceptScore W4386076172C154945302 @default.
- W4386076172 hasConceptScore W4386076172C177264268 @default.
- W4386076172 hasConceptScore W4386076172C186767784 @default.
- W4386076172 hasConceptScore W4386076172C199360897 @default.
- W4386076172 hasConceptScore W4386076172C202444582 @default.
- W4386076172 hasConceptScore W4386076172C33704608 @default.
- W4386076172 hasConceptScore W4386076172C33923547 @default.
- W4386076172 hasConceptScore W4386076172C41008148 @default.
- W4386076172 hasConceptScore W4386076172C73555534 @default.
- W4386076172 hasConceptScore W4386076172C94641424 @default.
- W4386076172 hasConceptScore W4386076172C9652623 @default.
- W4386076172 hasFunder F4320332999 @default.
- W4386076172 hasLocation W43860761721 @default.
- W4386076172 hasOpenAccess W4386076172 @default.