Matches in SemOpenAlex for { <https://semopenalex.org/work/W4386337492> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W4386337492 endingPage "339" @default.
- W4386337492 startingPage "322" @default.
- W4386337492 abstract "This paper describes a simple memory augmentation technique that employs tabular Q-learning to solve binary cell structured mazes with exits generated randomly at the start of each solution attempt. A standard tabular Q-learning can solve any maze with continuous learning; however, if the learning is stopped and the policy is frozen, the agent will not adapt to solve newly generated exits. To avoid using Recurrent Neural Networks RNNs to solve memory-required tasks, we designed and implemented a simple external memory to remember the agent’s cell visit history. This memory also expands the state information to hold more information, assisting tabular Q-learning in distinguishing its path from entering and exiting a maze corridor. Experiments on five maze problems of varying complexity are presented. The maze has two and four predefined exits; the exit will be randomly assigned at the start of each solution attempt. The results show that tabular Q-learning with a frozen policy can outperform standard deep-learning algorithms without incorporating RNNs into the model structure." @default.
- W4386337492 created "2023-09-01" @default.
- W4386337492 creator A5033216620 @default.
- W4386337492 creator A5074388845 @default.
- W4386337492 creator A5091964796 @default.
- W4386337492 date "2023-01-01" @default.
- W4386337492 modified "2023-09-30" @default.
- W4386337492 title "Finding Eulerian Tours in Mazes Using a Memory-Augmented Fixed Policy Function" @default.
- W4386337492 cites W1498436455 @default.
- W4386337492 cites W1662842982 @default.
- W4386337492 cites W2064746270 @default.
- W4386337492 cites W2069877466 @default.
- W4386337492 cites W2123491406 @default.
- W4386337492 cites W2144173369 @default.
- W4386337492 cites W2145339207 @default.
- W4386337492 cites W2153602280 @default.
- W4386337492 cites W2574473771 @default.
- W4386337492 cites W2981925632 @default.
- W4386337492 cites W3013886221 @default.
- W4386337492 cites W4200140111 @default.
- W4386337492 cites W4231410305 @default.
- W4386337492 doi "https://doi.org/10.1007/978-3-031-37717-4_22" @default.
- W4386337492 hasPublicationYear "2023" @default.
- W4386337492 type Work @default.
- W4386337492 citedByCount "0" @default.
- W4386337492 crossrefType "book-chapter" @default.
- W4386337492 hasAuthorship W4386337492A5033216620 @default.
- W4386337492 hasAuthorship W4386337492A5074388845 @default.
- W4386337492 hasAuthorship W4386337492A5091964796 @default.
- W4386337492 hasConcept C111472728 @default.
- W4386337492 hasConcept C119857082 @default.
- W4386337492 hasConcept C138885662 @default.
- W4386337492 hasConcept C14036430 @default.
- W4386337492 hasConcept C147168706 @default.
- W4386337492 hasConcept C154945302 @default.
- W4386337492 hasConcept C199360897 @default.
- W4386337492 hasConcept C2777735758 @default.
- W4386337492 hasConcept C2780586882 @default.
- W4386337492 hasConcept C41008148 @default.
- W4386337492 hasConcept C50644808 @default.
- W4386337492 hasConcept C78458016 @default.
- W4386337492 hasConcept C86803240 @default.
- W4386337492 hasConceptScore W4386337492C111472728 @default.
- W4386337492 hasConceptScore W4386337492C119857082 @default.
- W4386337492 hasConceptScore W4386337492C138885662 @default.
- W4386337492 hasConceptScore W4386337492C14036430 @default.
- W4386337492 hasConceptScore W4386337492C147168706 @default.
- W4386337492 hasConceptScore W4386337492C154945302 @default.
- W4386337492 hasConceptScore W4386337492C199360897 @default.
- W4386337492 hasConceptScore W4386337492C2777735758 @default.
- W4386337492 hasConceptScore W4386337492C2780586882 @default.
- W4386337492 hasConceptScore W4386337492C41008148 @default.
- W4386337492 hasConceptScore W4386337492C50644808 @default.
- W4386337492 hasConceptScore W4386337492C78458016 @default.
- W4386337492 hasConceptScore W4386337492C86803240 @default.
- W4386337492 hasLocation W43863374921 @default.
- W4386337492 hasOpenAccess W4386337492 @default.
- W4386337492 hasPrimaryLocation W43863374921 @default.
- W4386337492 hasRelatedWork W2902723393 @default.
- W4386337492 hasRelatedWork W2961085424 @default.
- W4386337492 hasRelatedWork W3046775127 @default.
- W4386337492 hasRelatedWork W4281386417 @default.
- W4386337492 hasRelatedWork W4285260836 @default.
- W4386337492 hasRelatedWork W4286629047 @default.
- W4386337492 hasRelatedWork W4306321456 @default.
- W4386337492 hasRelatedWork W4306674287 @default.
- W4386337492 hasRelatedWork W4327831767 @default.
- W4386337492 hasRelatedWork W4224009465 @default.
- W4386337492 isParatext "false" @default.
- W4386337492 isRetracted "false" @default.
- W4386337492 workType "book-chapter" @default.