Matches in SemOpenAlex for { <https://semopenalex.org/work/W4300574580> ?p ?o ?g. }
Showing items 1 to 75 of
75
with 100 items per page.
- W4300574580 abstract "We propose Episodic Backward Update (EBU) - a novel deep reinforcement learning algorithm with a direct value propagation. In contrast to the conventional use of the experience replay with uniform random sampling, our agent samples a whole episode and successively propagates the value of a state to its previous states. Our computationally efficient recursive algorithm allows sparse and delayed rewards to propagate directly through all transitions of the sampled episode. We theoretically prove the convergence of the EBU method and experimentally demonstrate its performance in both deterministic and stochastic environments. Especially in 49 games of Atari 2600 domain, EBU achieves the same mean and median human normalized performance of DQN by using only 5% and 10% of samples, respectively." @default.
- W4300574580 created "2022-10-03" @default.
- W4300574580 creator A5041710947 @default.
- W4300574580 creator A5057794084 @default.
- W4300574580 creator A5085845584 @default.
- W4300574580 date "2018-05-31" @default.
- W4300574580 modified "2023-10-16" @default.
- W4300574580 title "Sample-Efficient Deep Reinforcement Learning via Episodic Backward Update" @default.
- W4300574580 doi "https://doi.org/10.48550/arxiv.1805.12375" @default.
- W4300574580 hasPublicationYear "2018" @default.
- W4300574580 type Work @default.
- W4300574580 citedByCount "0" @default.
- W4300574580 crossrefType "posted-content" @default.
- W4300574580 hasAuthorship W4300574580A5041710947 @default.
- W4300574580 hasAuthorship W4300574580A5057794084 @default.
- W4300574580 hasAuthorship W4300574580A5085845584 @default.
- W4300574580 hasBestOaLocation W43005745801 @default.
- W4300574580 hasConcept C106131492 @default.
- W4300574580 hasConcept C11413529 @default.
- W4300574580 hasConcept C119857082 @default.
- W4300574580 hasConcept C126255220 @default.
- W4300574580 hasConcept C134306372 @default.
- W4300574580 hasConcept C140779682 @default.
- W4300574580 hasConcept C154945302 @default.
- W4300574580 hasConcept C162324750 @default.
- W4300574580 hasConcept C185592680 @default.
- W4300574580 hasConcept C198531522 @default.
- W4300574580 hasConcept C2776291640 @default.
- W4300574580 hasConcept C2776502983 @default.
- W4300574580 hasConcept C2777303404 @default.
- W4300574580 hasConcept C31972630 @default.
- W4300574580 hasConcept C33923547 @default.
- W4300574580 hasConcept C36503486 @default.
- W4300574580 hasConcept C41008148 @default.
- W4300574580 hasConcept C43617362 @default.
- W4300574580 hasConcept C48103436 @default.
- W4300574580 hasConcept C50522688 @default.
- W4300574580 hasConcept C97541855 @default.
- W4300574580 hasConceptScore W4300574580C106131492 @default.
- W4300574580 hasConceptScore W4300574580C11413529 @default.
- W4300574580 hasConceptScore W4300574580C119857082 @default.
- W4300574580 hasConceptScore W4300574580C126255220 @default.
- W4300574580 hasConceptScore W4300574580C134306372 @default.
- W4300574580 hasConceptScore W4300574580C140779682 @default.
- W4300574580 hasConceptScore W4300574580C154945302 @default.
- W4300574580 hasConceptScore W4300574580C162324750 @default.
- W4300574580 hasConceptScore W4300574580C185592680 @default.
- W4300574580 hasConceptScore W4300574580C198531522 @default.
- W4300574580 hasConceptScore W4300574580C2776291640 @default.
- W4300574580 hasConceptScore W4300574580C2776502983 @default.
- W4300574580 hasConceptScore W4300574580C2777303404 @default.
- W4300574580 hasConceptScore W4300574580C31972630 @default.
- W4300574580 hasConceptScore W4300574580C33923547 @default.
- W4300574580 hasConceptScore W4300574580C36503486 @default.
- W4300574580 hasConceptScore W4300574580C41008148 @default.
- W4300574580 hasConceptScore W4300574580C43617362 @default.
- W4300574580 hasConceptScore W4300574580C48103436 @default.
- W4300574580 hasConceptScore W4300574580C50522688 @default.
- W4300574580 hasConceptScore W4300574580C97541855 @default.
- W4300574580 hasLocation W43005745801 @default.
- W4300574580 hasOpenAccess W4300574580 @default.
- W4300574580 hasPrimaryLocation W43005745801 @default.
- W4300574580 hasRelatedWork W2101964129 @default.
- W4300574580 hasRelatedWork W2923653485 @default.
- W4300574580 hasRelatedWork W2952472710 @default.
- W4300574580 hasRelatedWork W2957776456 @default.
- W4300574580 hasRelatedWork W4206669594 @default.
- W4300574580 hasRelatedWork W4224287422 @default.
- W4300574580 hasRelatedWork W4255994452 @default.
- W4300574580 hasRelatedWork W4319773215 @default.
- W4300574580 hasRelatedWork W4361026739 @default.
- W4300574580 hasRelatedWork W4361253176 @default.
- W4300574580 isParatext "false" @default.
- W4300574580 isRetracted "false" @default.
- W4300574580 workType "article" @default.