Matches in SemOpenAlex for { <https://semopenalex.org/work/W4384346098> ?p ?o ?g. }
Showing items 1 to 98 of
98
with 100 items per page.
- W4384346098 endingPage "297" @default.
- W4384346098 startingPage "286" @default.
- W4384346098 abstract "Q-DPP is a multi-agent reinforcement learning (MARL) algorithm capable of eliminating the priori structural constraints on the central value function. However, due to the high-dimensional matrix operations with the increasing of the number of agents, Q-DPP shows poor empirical performance in complex environment, such as Starcraft Multi-Agent Challenge (SMAC). To improve the scalability and reduce computational complexity of Q-DPP, we propose a novel Deep Determinantal Q-Learning with Role Aware (RO-DPP). Concretely, we introduce the role concept by learning a role selector and cluster actions based on the impact of agent’s actions on the environment and other agents, and thus break down the joint action spaces into reduced role action spaces, which computes the kernel matrix of Q-DPP in a greatly reduced observation-action space. The experiments on micromanagement benchmark of famous game StarCraft II validate the effectiveness of the proposed RO-DPP." @default.
- W4384346098 created "2023-07-15" @default.
- W4384346098 creator A5021549646 @default.
- W4384346098 creator A5056485902 @default.
- W4384346098 creator A5071037763 @default.
- W4384346098 creator A5075095237 @default.
- W4384346098 date "2023-01-01" @default.
- W4384346098 modified "2023-09-24" @default.
- W4384346098 title "Deep Determinantal Q-Learning with Role Aware" @default.
- W4384346098 cites W1497040579 @default.
- W4384346098 cites W1641379095 @default.
- W4384346098 cites W2546571074 @default.
- W4384346098 cites W2617547828 @default.
- W4384346098 cites W2747213132 @default.
- W4384346098 cites W2963871073 @default.
- W4384346098 cites W3100944043 @default.
- W4384346098 cites W3104860527 @default.
- W4384346098 doi "https://doi.org/10.1007/978-3-031-36822-6_25" @default.
- W4384346098 hasPublicationYear "2023" @default.
- W4384346098 type Work @default.
- W4384346098 citedByCount "0" @default.
- W4384346098 crossrefType "book-chapter" @default.
- W4384346098 hasAuthorship W4384346098A5021549646 @default.
- W4384346098 hasAuthorship W4384346098A5056485902 @default.
- W4384346098 hasAuthorship W4384346098A5071037763 @default.
- W4384346098 hasAuthorship W4384346098A5075095237 @default.
- W4384346098 hasConcept C106487976 @default.
- W4384346098 hasConcept C111472728 @default.
- W4384346098 hasConcept C118615104 @default.
- W4384346098 hasConcept C121332964 @default.
- W4384346098 hasConcept C13280743 @default.
- W4384346098 hasConcept C138885662 @default.
- W4384346098 hasConcept C14036430 @default.
- W4384346098 hasConcept C154945302 @default.
- W4384346098 hasConcept C158693339 @default.
- W4384346098 hasConcept C159985019 @default.
- W4384346098 hasConcept C185798385 @default.
- W4384346098 hasConcept C192562407 @default.
- W4384346098 hasConcept C205649164 @default.
- W4384346098 hasConcept C2780791683 @default.
- W4384346098 hasConcept C33923547 @default.
- W4384346098 hasConcept C41008148 @default.
- W4384346098 hasConcept C48044578 @default.
- W4384346098 hasConcept C62520636 @default.
- W4384346098 hasConcept C64812099 @default.
- W4384346098 hasConcept C72010251 @default.
- W4384346098 hasConcept C74193536 @default.
- W4384346098 hasConcept C75553542 @default.
- W4384346098 hasConcept C77088390 @default.
- W4384346098 hasConcept C78458016 @default.
- W4384346098 hasConcept C80444323 @default.
- W4384346098 hasConcept C86803240 @default.
- W4384346098 hasConcept C97541855 @default.
- W4384346098 hasConceptScore W4384346098C106487976 @default.
- W4384346098 hasConceptScore W4384346098C111472728 @default.
- W4384346098 hasConceptScore W4384346098C118615104 @default.
- W4384346098 hasConceptScore W4384346098C121332964 @default.
- W4384346098 hasConceptScore W4384346098C13280743 @default.
- W4384346098 hasConceptScore W4384346098C138885662 @default.
- W4384346098 hasConceptScore W4384346098C14036430 @default.
- W4384346098 hasConceptScore W4384346098C154945302 @default.
- W4384346098 hasConceptScore W4384346098C158693339 @default.
- W4384346098 hasConceptScore W4384346098C159985019 @default.
- W4384346098 hasConceptScore W4384346098C185798385 @default.
- W4384346098 hasConceptScore W4384346098C192562407 @default.
- W4384346098 hasConceptScore W4384346098C205649164 @default.
- W4384346098 hasConceptScore W4384346098C2780791683 @default.
- W4384346098 hasConceptScore W4384346098C33923547 @default.
- W4384346098 hasConceptScore W4384346098C41008148 @default.
- W4384346098 hasConceptScore W4384346098C48044578 @default.
- W4384346098 hasConceptScore W4384346098C62520636 @default.
- W4384346098 hasConceptScore W4384346098C64812099 @default.
- W4384346098 hasConceptScore W4384346098C72010251 @default.
- W4384346098 hasConceptScore W4384346098C74193536 @default.
- W4384346098 hasConceptScore W4384346098C75553542 @default.
- W4384346098 hasConceptScore W4384346098C77088390 @default.
- W4384346098 hasConceptScore W4384346098C78458016 @default.
- W4384346098 hasConceptScore W4384346098C80444323 @default.
- W4384346098 hasConceptScore W4384346098C86803240 @default.
- W4384346098 hasConceptScore W4384346098C97541855 @default.
- W4384346098 hasLocation W43843460981 @default.
- W4384346098 hasOpenAccess W4384346098 @default.
- W4384346098 hasPrimaryLocation W43843460981 @default.
- W4384346098 hasRelatedWork W112744582 @default.
- W4384346098 hasRelatedWork W1485630101 @default.
- W4384346098 hasRelatedWork W1666765134 @default.
- W4384346098 hasRelatedWork W1992807924 @default.
- W4384346098 hasRelatedWork W2151702863 @default.
- W4384346098 hasRelatedWork W2461970972 @default.
- W4384346098 hasRelatedWork W2789601449 @default.
- W4384346098 hasRelatedWork W2910876866 @default.
- W4384346098 hasRelatedWork W4296474751 @default.
- W4384346098 hasRelatedWork W4301846872 @default.
- W4384346098 isParatext "false" @default.
- W4384346098 isRetracted "false" @default.
- W4384346098 workType "book-chapter" @default.