Matches in SemOpenAlex for { <https://semopenalex.org/work/W3202114911> ?p ?o ?g. }
- W3202114911 abstract "Off-policy Actor-Critic algorithms have demonstrated phenomenal experimental performance but still require better explanations. To this end, we show its policy evaluation error on the distribution of transitions decomposes into: a Bellman error, a bias from policy mismatch, and a variance term from sampling. By comparing the magnitude of bias and variance, we explain the success of the Emphasizing Recent Experience sampling and 1/age weighted sampling. Both sampling strategies yield smaller bias and variance and are hence preferable to uniform sampling." @default.
- W3202114911 created "2021-10-11" @default.
- W3202114911 creator A5041274293 @default.
- W3202114911 creator A5085851733 @default.
- W3202114911 date "2021-10-05" @default.
- W3202114911 modified "2023-09-27" @default.
- W3202114911 title "Explaining Off-Policy Actor-Critic From A Bias-Variance Perspective" @default.
- W3202114911 cites W1514587017 @default.
- W3202114911 cites W1771410628 @default.
- W3202114911 cites W2015731569 @default.
- W3202114911 cites W2141559645 @default.
- W3202114911 cites W2155027007 @default.
- W3202114911 cites W2158782408 @default.
- W3202114911 cites W2165150801 @default.
- W3202114911 cites W2736601468 @default.
- W3202114911 cites W2890022552 @default.
- W3202114911 cites W2948708918 @default.
- W3202114911 cites W2962734844 @default.
- W3202114911 cites W2962802563 @default.
- W3202114911 cites W2962879692 @default.
- W3202114911 cites W2962902376 @default.
- W3202114911 cites W2963215512 @default.
- W3202114911 cites W2963296584 @default.
- W3202114911 cites W2963477884 @default.
- W3202114911 cites W2963733916 @default.
- W3202114911 cites W2963836885 @default.
- W3202114911 cites W2963923407 @default.
- W3202114911 cites W2964068481 @default.
- W3202114911 cites W2964279789 @default.
- W3202114911 cites W2964297722 @default.
- W3202114911 cites W2991522342 @default.
- W3202114911 cites W2992545808 @default.
- W3202114911 cites W2998562709 @default.
- W3202114911 cites W3022566517 @default.
- W3202114911 cites W3034541690 @default.
- W3202114911 cites W3035157451 @default.
- W3202114911 cites W3035954878 @default.
- W3202114911 cites W3037792667 @default.
- W3202114911 cites W3046395471 @default.
- W3202114911 cites W3098593577 @default.
- W3202114911 cites W3101192004 @default.
- W3202114911 cites W3102827805 @default.
- W3202114911 cites W3104599378 @default.
- W3202114911 cites W3105602917 @default.
- W3202114911 cites W3115291921 @default.
- W3202114911 cites W3157615692 @default.
- W3202114911 cites W3159738529 @default.
- W3202114911 doi "https://doi.org/10.48550/arxiv.2110.02421" @default.
- W3202114911 hasPublicationYear "2021" @default.
- W3202114911 type Work @default.
- W3202114911 sameAs 3202114911 @default.
- W3202114911 citedByCount "0" @default.
- W3202114911 crossrefType "posted-content" @default.
- W3202114911 hasAuthorship W3202114911A5041274293 @default.
- W3202114911 hasAuthorship W3202114911A5085851733 @default.
- W3202114911 hasBestOaLocation W32021149111 @default.
- W3202114911 hasConcept C105795698 @default.
- W3202114911 hasConcept C106131492 @default.
- W3202114911 hasConcept C121332964 @default.
- W3202114911 hasConcept C121955636 @default.
- W3202114911 hasConcept C12713177 @default.
- W3202114911 hasConcept C129848803 @default.
- W3202114911 hasConcept C140779682 @default.
- W3202114911 hasConcept C149782125 @default.
- W3202114911 hasConcept C154945302 @default.
- W3202114911 hasConcept C162324750 @default.
- W3202114911 hasConcept C165473641 @default.
- W3202114911 hasConcept C19499675 @default.
- W3202114911 hasConcept C196083921 @default.
- W3202114911 hasConcept C19619285 @default.
- W3202114911 hasConcept C31972630 @default.
- W3202114911 hasConcept C33923547 @default.
- W3202114911 hasConcept C41008148 @default.
- W3202114911 hasConcept C52740198 @default.
- W3202114911 hasConcept C61797465 @default.
- W3202114911 hasConcept C62520636 @default.
- W3202114911 hasConcept C75917345 @default.
- W3202114911 hasConceptScore W3202114911C105795698 @default.
- W3202114911 hasConceptScore W3202114911C106131492 @default.
- W3202114911 hasConceptScore W3202114911C121332964 @default.
- W3202114911 hasConceptScore W3202114911C121955636 @default.
- W3202114911 hasConceptScore W3202114911C12713177 @default.
- W3202114911 hasConceptScore W3202114911C129848803 @default.
- W3202114911 hasConceptScore W3202114911C140779682 @default.
- W3202114911 hasConceptScore W3202114911C149782125 @default.
- W3202114911 hasConceptScore W3202114911C154945302 @default.
- W3202114911 hasConceptScore W3202114911C162324750 @default.
- W3202114911 hasConceptScore W3202114911C165473641 @default.
- W3202114911 hasConceptScore W3202114911C19499675 @default.
- W3202114911 hasConceptScore W3202114911C196083921 @default.
- W3202114911 hasConceptScore W3202114911C19619285 @default.
- W3202114911 hasConceptScore W3202114911C31972630 @default.
- W3202114911 hasConceptScore W3202114911C33923547 @default.
- W3202114911 hasConceptScore W3202114911C41008148 @default.
- W3202114911 hasConceptScore W3202114911C52740198 @default.
- W3202114911 hasConceptScore W3202114911C61797465 @default.
- W3202114911 hasConceptScore W3202114911C62520636 @default.
- W3202114911 hasConceptScore W3202114911C75917345 @default.
- W3202114911 hasLocation W32021149111 @default.
- W3202114911 hasOpenAccess W3202114911 @default.