Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386076671> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W4386076671 abstract "Learning representations for individual instances when only bag-level labels are available is a fundamental challenge in multiple instance learning (MIL). Recent works have shown promising results using contrastive self-supervised learning (CSSL), which learns to push apart representations corresponding to two different randomly-selected instances. Unfortunately, in real-world applications such as medical image classification, there is often class imbalance, so randomly-selected instances mostly belong to the same majority class, which precludes CSSL from learning inter-class differences. To address this issue, we propose a novel framework, Iterative Self-paced Supervised Contrastive Learning for MIL Representations (ItS2CLR), which improves the learned representation by exploiting instance-level pseudo labels derived from the bag-level labels. The framework employs a novel self-paced sampling strategy to ensure the accuracy of pseudo labels. We evaluate ItS2CLR on three medical datasets, showing that it improves the quality of instance-level pseudo labels and representations, and outperforms existing MIL methods in terms of both bag and instance level accuracy. <sup xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>1</sup> <sup xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>1</sup> Code is available at https://github.com/Kangningthu/ItS2CLR" @default.
- W4386076671 created "2023-08-23" @default.
- W4386076671 creator A5001916762 @default.
- W4386076671 creator A5003735451 @default.
- W4386076671 creator A5010954824 @default.
- W4386076671 creator A5034482431 @default.
- W4386076671 creator A5044336556 @default.
- W4386076671 creator A5066499149 @default.
- W4386076671 creator A5079997593 @default.
- W4386076671 date "2023-06-01" @default.
- W4386076671 modified "2023-10-06" @default.
- W4386076671 title "Multiple Instance Learning via Iterative Self-Paced Supervised Contrastive Learning" @default.
- W4386076671 cites W1945608308 @default.
- W4386076671 cites W2108598243 @default.
- W4386076671 cites W2194775991 @default.
- W4386076671 cites W2343160907 @default.
- W4386076671 cites W2760946358 @default.
- W4386076671 cites W2772723798 @default.
- W4386076671 cites W2921491036 @default.
- W4386076671 cites W2939176882 @default.
- W4386076671 cites W2956228567 @default.
- W4386076671 cites W2986297814 @default.
- W4386076671 cites W3012095418 @default.
- W4386076671 cites W3043835773 @default.
- W4386076671 cites W3093416812 @default.
- W4386076671 cites W3109976102 @default.
- W4386076671 cites W3112635222 @default.
- W4386076671 cites W3120430728 @default.
- W4386076671 cites W3135547872 @default.
- W4386076671 cites W3154677174 @default.
- W4386076671 cites W3158714121 @default.
- W4386076671 cites W3159481202 @default.
- W4386076671 cites W3176719058 @default.
- W4386076671 cites W3195767702 @default.
- W4386076671 cites W3201701662 @default.
- W4386076671 cites W3212889265 @default.
- W4386076671 cites W4214481550 @default.
- W4386076671 cites W4286908677 @default.
- W4386076671 cites W4293861233 @default.
- W4386076671 cites W4312765357 @default.
- W4386076671 doi "https://doi.org/10.1109/cvpr52729.2023.00327" @default.
- W4386076671 hasPublicationYear "2023" @default.
- W4386076671 type Work @default.
- W4386076671 citedByCount "1" @default.
- W4386076671 crossrefType "proceedings-article" @default.
- W4386076671 hasAuthorship W4386076671A5001916762 @default.
- W4386076671 hasAuthorship W4386076671A5003735451 @default.
- W4386076671 hasAuthorship W4386076671A5010954824 @default.
- W4386076671 hasAuthorship W4386076671A5034482431 @default.
- W4386076671 hasAuthorship W4386076671A5044336556 @default.
- W4386076671 hasAuthorship W4386076671A5066499149 @default.
- W4386076671 hasAuthorship W4386076671A5079997593 @default.
- W4386076671 hasConcept C119857082 @default.
- W4386076671 hasConcept C154945302 @default.
- W4386076671 hasConcept C177264268 @default.
- W4386076671 hasConcept C17744445 @default.
- W4386076671 hasConcept C199360897 @default.
- W4386076671 hasConcept C199539241 @default.
- W4386076671 hasConcept C204321447 @default.
- W4386076671 hasConcept C2776359362 @default.
- W4386076671 hasConcept C2776760102 @default.
- W4386076671 hasConcept C2777212361 @default.
- W4386076671 hasConcept C41008148 @default.
- W4386076671 hasConcept C94625758 @default.
- W4386076671 hasConceptScore W4386076671C119857082 @default.
- W4386076671 hasConceptScore W4386076671C154945302 @default.
- W4386076671 hasConceptScore W4386076671C177264268 @default.
- W4386076671 hasConceptScore W4386076671C17744445 @default.
- W4386076671 hasConceptScore W4386076671C199360897 @default.
- W4386076671 hasConceptScore W4386076671C199539241 @default.
- W4386076671 hasConceptScore W4386076671C204321447 @default.
- W4386076671 hasConceptScore W4386076671C2776359362 @default.
- W4386076671 hasConceptScore W4386076671C2776760102 @default.
- W4386076671 hasConceptScore W4386076671C2777212361 @default.
- W4386076671 hasConceptScore W4386076671C41008148 @default.
- W4386076671 hasConceptScore W4386076671C94625758 @default.
- W4386076671 hasLocation W43860766711 @default.
- W4386076671 hasOpenAccess W4386076671 @default.
- W4386076671 hasPrimaryLocation W43860766711 @default.
- W4386076671 hasRelatedWork W2961085424 @default.
- W4386076671 hasRelatedWork W3046775127 @default.
- W4386076671 hasRelatedWork W3170094116 @default.
- W4386076671 hasRelatedWork W4205958290 @default.
- W4386076671 hasRelatedWork W4285260836 @default.
- W4386076671 hasRelatedWork W4286629047 @default.
- W4386076671 hasRelatedWork W4306321456 @default.
- W4386076671 hasRelatedWork W4306674287 @default.
- W4386076671 hasRelatedWork W4386462264 @default.
- W4386076671 hasRelatedWork W4224009465 @default.
- W4386076671 isParatext "false" @default.
- W4386076671 isRetracted "false" @default.
- W4386076671 workType "article" @default.