Matches in SemOpenAlex for { <https://semopenalex.org/work/W2126330148> ?p ?o ?g. }
Showing items 1 to 90 of
90
with 100 items per page.
- W2126330148 abstract "Importance sampling (IS) and actor-critic are two methods which have been used to reduce the variance of gradient estimates in policy gradient optimization methods. We show how IS can be used with temporal difference methods to estimate a cost function parameter for one policy using the entire history of system interactions incorporating many different policies. The resulting algorithm is then applied to improving gradient estimates in a policy gradient optimization. The empirical results demonstrate a 20-40 times reduction in variance over the IS estimator for an example queueing problem, resulting in a similar factor of improvement in convergence for a gradient search" @default.
- W2126330148 created "2016-06-24" @default.
- W2126330148 creator A5044409973 @default.
- W2126330148 creator A5076006629 @default.
- W2126330148 creator A5091608832 @default.
- W2126330148 date "2006-01-01" @default.
- W2126330148 modified "2023-10-08" @default.
- W2126330148 title "Importance sampling actor-critic algorithms" @default.
- W2126330148 cites W1507222174 @default.
- W2126330148 cites W2009303086 @default.
- W2126330148 cites W2139418546 @default.
- W2126330148 doi "https://doi.org/10.1109/acc.2006.1656451" @default.
- W2126330148 hasPublicationYear "2006" @default.
- W2126330148 type Work @default.
- W2126330148 sameAs 2126330148 @default.
- W2126330148 citedByCount "7" @default.
- W2126330148 countsByYear W21263301482012 @default.
- W2126330148 countsByYear W21263301482015 @default.
- W2126330148 countsByYear W21263301482017 @default.
- W2126330148 crossrefType "proceedings-article" @default.
- W2126330148 hasAuthorship W2126330148A5044409973 @default.
- W2126330148 hasAuthorship W2126330148A5076006629 @default.
- W2126330148 hasAuthorship W2126330148A5091608832 @default.
- W2126330148 hasBestOaLocation W21263301482 @default.
- W2126330148 hasConcept C105795698 @default.
- W2126330148 hasConcept C106131492 @default.
- W2126330148 hasConcept C11413529 @default.
- W2126330148 hasConcept C115680565 @default.
- W2126330148 hasConcept C121955636 @default.
- W2126330148 hasConcept C126255220 @default.
- W2126330148 hasConcept C14036430 @default.
- W2126330148 hasConcept C140779682 @default.
- W2126330148 hasConcept C144133560 @default.
- W2126330148 hasConcept C162324750 @default.
- W2126330148 hasConcept C185429906 @default.
- W2126330148 hasConcept C19499675 @default.
- W2126330148 hasConcept C196083921 @default.
- W2126330148 hasConcept C22684755 @default.
- W2126330148 hasConcept C2777303404 @default.
- W2126330148 hasConcept C31258907 @default.
- W2126330148 hasConcept C31972630 @default.
- W2126330148 hasConcept C33923547 @default.
- W2126330148 hasConcept C41008148 @default.
- W2126330148 hasConcept C50522688 @default.
- W2126330148 hasConcept C52740198 @default.
- W2126330148 hasConcept C62644790 @default.
- W2126330148 hasConcept C78458016 @default.
- W2126330148 hasConcept C86803240 @default.
- W2126330148 hasConceptScore W2126330148C105795698 @default.
- W2126330148 hasConceptScore W2126330148C106131492 @default.
- W2126330148 hasConceptScore W2126330148C11413529 @default.
- W2126330148 hasConceptScore W2126330148C115680565 @default.
- W2126330148 hasConceptScore W2126330148C121955636 @default.
- W2126330148 hasConceptScore W2126330148C126255220 @default.
- W2126330148 hasConceptScore W2126330148C14036430 @default.
- W2126330148 hasConceptScore W2126330148C140779682 @default.
- W2126330148 hasConceptScore W2126330148C144133560 @default.
- W2126330148 hasConceptScore W2126330148C162324750 @default.
- W2126330148 hasConceptScore W2126330148C185429906 @default.
- W2126330148 hasConceptScore W2126330148C19499675 @default.
- W2126330148 hasConceptScore W2126330148C196083921 @default.
- W2126330148 hasConceptScore W2126330148C22684755 @default.
- W2126330148 hasConceptScore W2126330148C2777303404 @default.
- W2126330148 hasConceptScore W2126330148C31258907 @default.
- W2126330148 hasConceptScore W2126330148C31972630 @default.
- W2126330148 hasConceptScore W2126330148C33923547 @default.
- W2126330148 hasConceptScore W2126330148C41008148 @default.
- W2126330148 hasConceptScore W2126330148C50522688 @default.
- W2126330148 hasConceptScore W2126330148C52740198 @default.
- W2126330148 hasConceptScore W2126330148C62644790 @default.
- W2126330148 hasConceptScore W2126330148C78458016 @default.
- W2126330148 hasConceptScore W2126330148C86803240 @default.
- W2126330148 hasLocation W21263301481 @default.
- W2126330148 hasLocation W21263301482 @default.
- W2126330148 hasOpenAccess W2126330148 @default.
- W2126330148 hasPrimaryLocation W21263301481 @default.
- W2126330148 hasRelatedWork W1977112355 @default.
- W2126330148 hasRelatedWork W2010899721 @default.
- W2126330148 hasRelatedWork W2039667440 @default.
- W2126330148 hasRelatedWork W2169367269 @default.
- W2126330148 hasRelatedWork W2298254442 @default.
- W2126330148 hasRelatedWork W2376074353 @default.
- W2126330148 hasRelatedWork W2900543860 @default.
- W2126330148 hasRelatedWork W2980964960 @default.
- W2126330148 hasRelatedWork W3031816500 @default.
- W2126330148 hasRelatedWork W3034780828 @default.
- W2126330148 isParatext "false" @default.
- W2126330148 isRetracted "false" @default.
- W2126330148 magId "2126330148" @default.
- W2126330148 workType "article" @default.