Matches in SemOpenAlex for { <https://semopenalex.org/work/W4363676058> ?p ?o ?g. }
Showing items 1 to 79 of
79
with 100 items per page.
- W4363676058 abstract "We extend the provably convergent Full Gradient DQN algorithm for discounted reward Markov decision processes from Avrachenkov et al. (2021) to average reward problems. We experimentally compare widely used RVI Q-Learning with recently proposed Differential Q-Learning in the neural function approximation setting with Full Gradient DQN and DQN. We also extend this to learn Whittle indices for Markovian restless multi-armed bandits. We observe a better convergence rate of the proposed Full Gradient variant across different tasks." @default.
- W4363676058 created "2023-04-11" @default.
- W4363676058 creator A5018541798 @default.
- W4363676058 creator A5027522398 @default.
- W4363676058 creator A5047001818 @default.
- W4363676058 date "2023-04-07" @default.
- W4363676058 modified "2023-09-27" @default.
- W4363676058 title "Full Gradient Deep Reinforcement Learning for Average-Reward Criterion" @default.
- W4363676058 doi "https://doi.org/10.48550/arxiv.2304.03729" @default.
- W4363676058 hasPublicationYear "2023" @default.
- W4363676058 type Work @default.
- W4363676058 citedByCount "0" @default.
- W4363676058 crossrefType "posted-content" @default.
- W4363676058 hasAuthorship W4363676058A5018541798 @default.
- W4363676058 hasAuthorship W4363676058A5027522398 @default.
- W4363676058 hasAuthorship W4363676058A5047001818 @default.
- W4363676058 hasBestOaLocation W43636760581 @default.
- W4363676058 hasConcept C105795698 @default.
- W4363676058 hasConcept C106189395 @default.
- W4363676058 hasConcept C115680565 @default.
- W4363676058 hasConcept C119857082 @default.
- W4363676058 hasConcept C126255220 @default.
- W4363676058 hasConcept C127413603 @default.
- W4363676058 hasConcept C14036430 @default.
- W4363676058 hasConcept C146978453 @default.
- W4363676058 hasConcept C153258448 @default.
- W4363676058 hasConcept C154945302 @default.
- W4363676058 hasConcept C159886148 @default.
- W4363676058 hasConcept C162324750 @default.
- W4363676058 hasConcept C2777303404 @default.
- W4363676058 hasConcept C28826006 @default.
- W4363676058 hasConcept C33923547 @default.
- W4363676058 hasConcept C41008148 @default.
- W4363676058 hasConcept C50522688 @default.
- W4363676058 hasConcept C50644808 @default.
- W4363676058 hasConcept C78458016 @default.
- W4363676058 hasConcept C86803240 @default.
- W4363676058 hasConcept C93226319 @default.
- W4363676058 hasConcept C97541855 @default.
- W4363676058 hasConcept C98763669 @default.
- W4363676058 hasConceptScore W4363676058C105795698 @default.
- W4363676058 hasConceptScore W4363676058C106189395 @default.
- W4363676058 hasConceptScore W4363676058C115680565 @default.
- W4363676058 hasConceptScore W4363676058C119857082 @default.
- W4363676058 hasConceptScore W4363676058C126255220 @default.
- W4363676058 hasConceptScore W4363676058C127413603 @default.
- W4363676058 hasConceptScore W4363676058C14036430 @default.
- W4363676058 hasConceptScore W4363676058C146978453 @default.
- W4363676058 hasConceptScore W4363676058C153258448 @default.
- W4363676058 hasConceptScore W4363676058C154945302 @default.
- W4363676058 hasConceptScore W4363676058C159886148 @default.
- W4363676058 hasConceptScore W4363676058C162324750 @default.
- W4363676058 hasConceptScore W4363676058C2777303404 @default.
- W4363676058 hasConceptScore W4363676058C28826006 @default.
- W4363676058 hasConceptScore W4363676058C33923547 @default.
- W4363676058 hasConceptScore W4363676058C41008148 @default.
- W4363676058 hasConceptScore W4363676058C50522688 @default.
- W4363676058 hasConceptScore W4363676058C50644808 @default.
- W4363676058 hasConceptScore W4363676058C78458016 @default.
- W4363676058 hasConceptScore W4363676058C86803240 @default.
- W4363676058 hasConceptScore W4363676058C93226319 @default.
- W4363676058 hasConceptScore W4363676058C97541855 @default.
- W4363676058 hasConceptScore W4363676058C98763669 @default.
- W4363676058 hasLocation W43636760581 @default.
- W4363676058 hasOpenAccess W4363676058 @default.
- W4363676058 hasPrimaryLocation W43636760581 @default.
- W4363676058 hasRelatedWork W1626977535 @default.
- W4363676058 hasRelatedWork W2029656683 @default.
- W4363676058 hasRelatedWork W2124144580 @default.
- W4363676058 hasRelatedWork W2128702080 @default.
- W4363676058 hasRelatedWork W2808418668 @default.
- W4363676058 hasRelatedWork W3140738360 @default.
- W4363676058 hasRelatedWork W3167472281 @default.
- W4363676058 hasRelatedWork W3198596521 @default.
- W4363676058 hasRelatedWork W4287640753 @default.
- W4363676058 hasRelatedWork W4312091648 @default.
- W4363676058 isParatext "false" @default.
- W4363676058 isRetracted "false" @default.
- W4363676058 workType "article" @default.