Matches in SemOpenAlex for { <https://semopenalex.org/work/W2963877897> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W2963877897 endingPage "497" @default.
- W2963877897 startingPage "489" @default.
- W2963877897 abstract "MART (Friedman, 2001, 2002), an ensemble model of boosted regression trees, is known to deliver high prediction accuracy for diverse tasks, and it is widely used in practice. However, it suffers an issue which we call over-specialization, wherein trees added at later iterations tend to impact the prediction of only a few instances, and make negligible contribution towards the remaining instances. This negatively affects the performance of the model on unseen data, and also makes the model over-sensitive to the contributions of the few, initially added tress. We show that the commonly used tool to address this issue, that of shrinkage, alleviates the problem only to a certain extent and the fundamental issue of over-specialization still remains. In this work, we explore a different approach to address the problem that of employing dropouts, a tool that has been recently proposed in the context of learning deep neural networks (Hinton et al., 2012). We propose a novel way of employing dropouts in MART, resulting in the DART algorithm. We evaluate DART on ranking, regression and classification tasks, using large scale, publicly available datasets, and show that DART outperforms MART in each of the tasks, with a significant margin. We also show that DART overcomes the issue of over-specialization to a considerable extent. Appearing in Proceedings of the 18 International Conference on Artificial Intelligence and Statistics (AISTATS) 2015, San Diego, CA, USA. JMLR: W&CP volume 38. Copyright 2015 by the authors." @default.
- W2963877897 created "2019-07-30" @default.
- W2963877897 creator A5036797884 @default.
- W2963877897 creator A5088004286 @default.
- W2963877897 date "2015-02-21" @default.
- W2963877897 modified "2023-09-24" @default.
- W2963877897 title "{DART: Dropouts meet Multiple Additive Regression Trees}" @default.
- W2963877897 cites W144691381 @default.
- W2963877897 cites W1678356000 @default.
- W2963877897 cites W1904365287 @default.
- W2963877897 cites W1931284523 @default.
- W2963877897 cites W2070493638 @default.
- W2963877897 cites W2115584760 @default.
- W2963877897 cites W2128877075 @default.
- W2963877897 cites W2129018774 @default.
- W2963877897 cites W2135695572 @default.
- W2963877897 cites W2143331230 @default.
- W2963877897 cites W2152722485 @default.
- W2963877897 cites W2884475480 @default.
- W2963877897 cites W35527955 @default.
- W2963877897 hasPublicationYear "2015" @default.
- W2963877897 type Work @default.
- W2963877897 sameAs 2963877897 @default.
- W2963877897 citedByCount "18" @default.
- W2963877897 countsByYear W29638778972017 @default.
- W2963877897 countsByYear W29638778972019 @default.
- W2963877897 countsByYear W29638778972020 @default.
- W2963877897 countsByYear W29638778972021 @default.
- W2963877897 crossrefType "proceedings-article" @default.
- W2963877897 hasAuthorship W2963877897A5036797884 @default.
- W2963877897 hasAuthorship W2963877897A5088004286 @default.
- W2963877897 hasConcept C105795698 @default.
- W2963877897 hasConcept C113174947 @default.
- W2963877897 hasConcept C119857082 @default.
- W2963877897 hasConcept C134306372 @default.
- W2963877897 hasConcept C151730666 @default.
- W2963877897 hasConcept C154945302 @default.
- W2963877897 hasConcept C189430467 @default.
- W2963877897 hasConcept C199360897 @default.
- W2963877897 hasConcept C2779343474 @default.
- W2963877897 hasConcept C2779417484 @default.
- W2963877897 hasConcept C33923547 @default.
- W2963877897 hasConcept C41008148 @default.
- W2963877897 hasConcept C50644808 @default.
- W2963877897 hasConcept C774472 @default.
- W2963877897 hasConcept C83546350 @default.
- W2963877897 hasConcept C86803240 @default.
- W2963877897 hasConceptScore W2963877897C105795698 @default.
- W2963877897 hasConceptScore W2963877897C113174947 @default.
- W2963877897 hasConceptScore W2963877897C119857082 @default.
- W2963877897 hasConceptScore W2963877897C134306372 @default.
- W2963877897 hasConceptScore W2963877897C151730666 @default.
- W2963877897 hasConceptScore W2963877897C154945302 @default.
- W2963877897 hasConceptScore W2963877897C189430467 @default.
- W2963877897 hasConceptScore W2963877897C199360897 @default.
- W2963877897 hasConceptScore W2963877897C2779343474 @default.
- W2963877897 hasConceptScore W2963877897C2779417484 @default.
- W2963877897 hasConceptScore W2963877897C33923547 @default.
- W2963877897 hasConceptScore W2963877897C41008148 @default.
- W2963877897 hasConceptScore W2963877897C50644808 @default.
- W2963877897 hasConceptScore W2963877897C774472 @default.
- W2963877897 hasConceptScore W2963877897C83546350 @default.
- W2963877897 hasConceptScore W2963877897C86803240 @default.
- W2963877897 hasLocation W29638778971 @default.
- W2963877897 hasOpenAccess W2963877897 @default.
- W2963877897 hasPrimaryLocation W29638778971 @default.
- W2963877897 hasRelatedWork W1517032226 @default.
- W2963877897 hasRelatedWork W1678356000 @default.
- W2963877897 hasRelatedWork W1775823180 @default.
- W2963877897 hasRelatedWork W1996348131 @default.
- W2963877897 hasRelatedWork W2024046085 @default.
- W2963877897 hasRelatedWork W2070493638 @default.
- W2963877897 hasRelatedWork W2091233897 @default.
- W2963877897 hasRelatedWork W2095705004 @default.
- W2963877897 hasRelatedWork W2097998348 @default.
- W2963877897 hasRelatedWork W2559898455 @default.
- W2963877897 hasRelatedWork W2768348081 @default.
- W2963877897 hasRelatedWork W2911964244 @default.
- W2963877897 hasRelatedWork W2912228103 @default.
- W2963877897 hasRelatedWork W2912934387 @default.
- W2963877897 hasRelatedWork W2955989140 @default.
- W2963877897 hasRelatedWork W3031484690 @default.
- W2963877897 hasRelatedWork W3088969596 @default.
- W2963877897 hasRelatedWork W3102476541 @default.
- W2963877897 hasRelatedWork W3122321081 @default.
- W2963877897 hasRelatedWork W98929787 @default.
- W2963877897 isParatext "false" @default.
- W2963877897 isRetracted "false" @default.
- W2963877897 magId "2963877897" @default.
- W2963877897 workType "article" @default.