Matches in SemOpenAlex for { <https://semopenalex.org/work/W3201631489> ?p ?o ?g. }
Showing items 1 to 75 of
75
with 100 items per page.
- W3201631489 abstract "This paper presents a methodology to exploit causation in deep reinforcement learning (DRL). We take advantage of a cognitive architecture that automatically decomposes the game world in proto-objects and records their position in 2D space across time. Therefore, while playing the game, proto-objects' locations define internal time sequences that can be compared with the reward sequence to select the proto-object that caused the reward. We propose a novel non-parametric information theoretic learning Granger causality (ITL-GC) estimator of directed information using Reny's entropy that is accurate in high dimensions. We integrate this module in a state-of-the-art DRL architecture (A3C) and show substantial improvement in the speed of convergence compared with conventional training." @default.
- W3201631489 created "2021-09-27" @default.
- W3201631489 creator A5019504861 @default.
- W3201631489 creator A5032451259 @default.
- W3201631489 date "2021-07-18" @default.
- W3201631489 modified "2023-09-23" @default.
- W3201631489 title "Speeding Up Reinforcement Learning by Exploiting Causality in Reward Sequences" @default.
- W3201631489 cites W2045509016 @default.
- W3201631489 cites W2096023955 @default.
- W3201631489 cites W2122882636 @default.
- W3201631489 cites W2133207235 @default.
- W3201631489 cites W2140911775 @default.
- W3201631489 cites W2148584298 @default.
- W3201631489 cites W2171866865 @default.
- W3201631489 cites W2178225550 @default.
- W3201631489 cites W2405642423 @default.
- W3201631489 cites W2786344118 @default.
- W3201631489 cites W2890967717 @default.
- W3201631489 cites W2962783375 @default.
- W3201631489 cites W2964043796 @default.
- W3201631489 cites W2964191931 @default.
- W3201631489 cites W2966030703 @default.
- W3201631489 cites W2971202257 @default.
- W3201631489 cites W3125933622 @default.
- W3201631489 doi "https://doi.org/10.1109/ijcnn52387.2021.9533910" @default.
- W3201631489 hasPublicationYear "2021" @default.
- W3201631489 type Work @default.
- W3201631489 sameAs 3201631489 @default.
- W3201631489 citedByCount "0" @default.
- W3201631489 crossrefType "proceedings-article" @default.
- W3201631489 hasAuthorship W3201631489A5019504861 @default.
- W3201631489 hasAuthorship W3201631489A5032451259 @default.
- W3201631489 hasConcept C106301342 @default.
- W3201631489 hasConcept C119857082 @default.
- W3201631489 hasConcept C121332964 @default.
- W3201631489 hasConcept C154945302 @default.
- W3201631489 hasConcept C162324750 @default.
- W3201631489 hasConcept C165696696 @default.
- W3201631489 hasConcept C2777303404 @default.
- W3201631489 hasConcept C2781238097 @default.
- W3201631489 hasConcept C38652104 @default.
- W3201631489 hasConcept C41008148 @default.
- W3201631489 hasConcept C50522688 @default.
- W3201631489 hasConcept C62520636 @default.
- W3201631489 hasConcept C97541855 @default.
- W3201631489 hasConceptScore W3201631489C106301342 @default.
- W3201631489 hasConceptScore W3201631489C119857082 @default.
- W3201631489 hasConceptScore W3201631489C121332964 @default.
- W3201631489 hasConceptScore W3201631489C154945302 @default.
- W3201631489 hasConceptScore W3201631489C162324750 @default.
- W3201631489 hasConceptScore W3201631489C165696696 @default.
- W3201631489 hasConceptScore W3201631489C2777303404 @default.
- W3201631489 hasConceptScore W3201631489C2781238097 @default.
- W3201631489 hasConceptScore W3201631489C38652104 @default.
- W3201631489 hasConceptScore W3201631489C41008148 @default.
- W3201631489 hasConceptScore W3201631489C50522688 @default.
- W3201631489 hasConceptScore W3201631489C62520636 @default.
- W3201631489 hasConceptScore W3201631489C97541855 @default.
- W3201631489 hasLocation W32016314891 @default.
- W3201631489 hasOpenAccess W3201631489 @default.
- W3201631489 hasPrimaryLocation W32016314891 @default.
- W3201631489 hasRelatedWork W1562959674 @default.
- W3201631489 hasRelatedWork W2331043530 @default.
- W3201631489 hasRelatedWork W2923653485 @default.
- W3201631489 hasRelatedWork W2952472710 @default.
- W3201631489 hasRelatedWork W2957776456 @default.
- W3201631489 hasRelatedWork W2964604098 @default.
- W3201631489 hasRelatedWork W2997512100 @default.
- W3201631489 hasRelatedWork W3022038857 @default.
- W3201631489 hasRelatedWork W4319083788 @default.
- W3201631489 hasRelatedWork W4361026739 @default.
- W3201631489 isParatext "false" @default.
- W3201631489 isRetracted "false" @default.
- W3201631489 magId "3201631489" @default.
- W3201631489 workType "article" @default.