Matches in SemOpenAlex for { <https://semopenalex.org/work/W2963482016> ?p ?o ?g. }
- W2963482016 abstract "We propose a new framework for designing estimators for off-policy evaluation in contextual bandits. Our approach is based on the asymptotically optimal doubly robust estimator, but we shrink the importance weights to minimize a bound on the mean squared error, which results in a better bias-variance tradeoff in finite samples. We use this optimization-based framework to obtain three estimators: (a) a weight-clipping estimator, (b) a new weight-shrinkage estimator, and (c) the first shrinkage-based estimator for combinatorial action sets. Extensive experiments in both standard and combinatorial bandit benchmark problems show that our estimators are highly adaptive and typically outperform state-of-the-art methods." @default.
- W2963482016 created "2019-07-30" @default.
- W2963482016 creator A5015082848 @default.
- W2963482016 creator A5068328730 @default.
- W2963482016 creator A5074724507 @default.
- W2963482016 creator A5089372170 @default.
- W2963482016 date "2019-07-22" @default.
- W2963482016 modified "2023-09-27" @default.
- W2963482016 title "Doubly robust off-policy evaluation with shrinkage" @default.
- W2963482016 cites W1094752974 @default.
- W2963482016 cites W1544839292 @default.
- W2963482016 cites W2012731843 @default.
- W2963482016 cites W2014373672 @default.
- W2963482016 cites W2079597004 @default.
- W2963482016 cites W2099471337 @default.
- W2963482016 cites W2113065326 @default.
- W2963482016 cites W2119850747 @default.
- W2963482016 cites W2120745256 @default.
- W2963482016 cites W2122124659 @default.
- W2963482016 cites W2137370054 @default.
- W2963482016 cites W2138909795 @default.
- W2963482016 cites W2188353343 @default.
- W2963482016 cites W2240609664 @default.
- W2963482016 cites W2471042628 @default.
- W2963482016 cites W2914156981 @default.
- W2963482016 cites W2962694783 @default.
- W2963482016 cites W2962736281 @default.
- W2963482016 cites W2962785510 @default.
- W2963482016 cites W2963323139 @default.
- W2963482016 cites W2964068481 @default.
- W2963482016 cites W2964233690 @default.
- W2963482016 cites W2964297722 @default.
- W2963482016 cites W3098679278 @default.
- W2963482016 cites W3122193054 @default.
- W2963482016 cites W3124983772 @default.
- W2963482016 cites W3131615941 @default.
- W2963482016 hasPublicationYear "2019" @default.
- W2963482016 type Work @default.
- W2963482016 sameAs 2963482016 @default.
- W2963482016 citedByCount "2" @default.
- W2963482016 countsByYear W29634820162020 @default.
- W2963482016 countsByYear W29634820162021 @default.
- W2963482016 crossrefType "posted-content" @default.
- W2963482016 hasAuthorship W2963482016A5015082848 @default.
- W2963482016 hasAuthorship W2963482016A5068328730 @default.
- W2963482016 hasAuthorship W2963482016A5074724507 @default.
- W2963482016 hasAuthorship W2963482016A5089372170 @default.
- W2963482016 hasConcept C102592046 @default.
- W2963482016 hasConcept C105795698 @default.
- W2963482016 hasConcept C121955636 @default.
- W2963482016 hasConcept C126255220 @default.
- W2963482016 hasConcept C13280743 @default.
- W2963482016 hasConcept C134306372 @default.
- W2963482016 hasConcept C139945424 @default.
- W2963482016 hasConcept C144133560 @default.
- W2963482016 hasConcept C165646398 @default.
- W2963482016 hasConcept C180145272 @default.
- W2963482016 hasConcept C185429906 @default.
- W2963482016 hasConcept C185798385 @default.
- W2963482016 hasConcept C196083921 @default.
- W2963482016 hasConcept C205649164 @default.
- W2963482016 hasConcept C33923547 @default.
- W2963482016 hasConcept C35594927 @default.
- W2963482016 hasConcept C41008148 @default.
- W2963482016 hasConcept C77553402 @default.
- W2963482016 hasConceptScore W2963482016C102592046 @default.
- W2963482016 hasConceptScore W2963482016C105795698 @default.
- W2963482016 hasConceptScore W2963482016C121955636 @default.
- W2963482016 hasConceptScore W2963482016C126255220 @default.
- W2963482016 hasConceptScore W2963482016C13280743 @default.
- W2963482016 hasConceptScore W2963482016C134306372 @default.
- W2963482016 hasConceptScore W2963482016C139945424 @default.
- W2963482016 hasConceptScore W2963482016C144133560 @default.
- W2963482016 hasConceptScore W2963482016C165646398 @default.
- W2963482016 hasConceptScore W2963482016C180145272 @default.
- W2963482016 hasConceptScore W2963482016C185429906 @default.
- W2963482016 hasConceptScore W2963482016C185798385 @default.
- W2963482016 hasConceptScore W2963482016C196083921 @default.
- W2963482016 hasConceptScore W2963482016C205649164 @default.
- W2963482016 hasConceptScore W2963482016C33923547 @default.
- W2963482016 hasConceptScore W2963482016C35594927 @default.
- W2963482016 hasConceptScore W2963482016C41008148 @default.
- W2963482016 hasConceptScore W2963482016C77553402 @default.
- W2963482016 hasLocation W29634820161 @default.
- W2963482016 hasOpenAccess W2963482016 @default.
- W2963482016 hasPrimaryLocation W29634820161 @default.
- W2963482016 hasRelatedWork W1481105389 @default.
- W2963482016 hasRelatedWork W1540138104 @default.
- W2963482016 hasRelatedWork W1546708447 @default.
- W2963482016 hasRelatedWork W1717817085 @default.
- W2963482016 hasRelatedWork W2023777598 @default.
- W2963482016 hasRelatedWork W2119653635 @default.
- W2963482016 hasRelatedWork W2133934706 @default.
- W2963482016 hasRelatedWork W2271643118 @default.
- W2963482016 hasRelatedWork W2605884907 @default.
- W2963482016 hasRelatedWork W2946421084 @default.
- W2963482016 hasRelatedWork W2949220983 @default.
- W2963482016 hasRelatedWork W3012270601 @default.
- W2963482016 hasRelatedWork W3015960270 @default.
- W2963482016 hasRelatedWork W3037491951 @default.