Matches in SemOpenAlex for { <https://semopenalex.org/work/W2969121442> ?p ?o ?g. }
- W2969121442 abstract "Multi-agent systems have a wide range of applications in cooperative and competitive tasks. As the number of agents increases, nonstationarity gets more serious in multi-agent reinforcement learning (MARL), which brings great difficulties to the learning process. Besides, current mainstream algorithms configure each agent an independent network,so that the memory usage increases linearly with the number of agents which greatly slows down the interaction with the environment. Inspired by Generative Adversarial Networks (GAN), this paper proposes an iterative update method (IU) to stabilize the nonstationary environment. Further, we add first-person perspective and represent all agents by only one network which can change agents' policies from sequential compute to batch compute. Similar to continual lifelong learning, we realize the iterative update method in this unified representative network (IUUR). In this method, iterative update can greatly alleviate the nonstationarity of the environment, unified representation can speed up the interaction with environment and avoid the linear growth of memory usage. Besides, this method does not bother decentralized execution and distributed deployment. Experiments show that compared with MADDPG, our algorithm achieves state-of-the-art performance and saves wall-clock time by a large margin especially with more agents." @default.
- W2969121442 created "2019-08-22" @default.
- W2969121442 creator A5062895972 @default.
- W2969121442 creator A5064045506 @default.
- W2969121442 creator A5079570262 @default.
- W2969121442 creator A5091771063 @default.
- W2969121442 date "2019-08-16" @default.
- W2969121442 modified "2023-09-27" @default.
- W2969121442 title "Iterative Update and Unified Representation for Multi-Agent Reinforcement Learning" @default.
- W2969121442 cites W1542941925 @default.
- W2969121442 cites W1564534945 @default.
- W2969121442 cites W1641379095 @default.
- W2969121442 cites W2007734734 @default.
- W2969121442 cites W206679605 @default.
- W2969121442 cites W2120846115 @default.
- W2969121442 cites W2121863487 @default.
- W2969121442 cites W2147492008 @default.
- W2969121442 cites W2173248099 @default.
- W2969121442 cites W2438667436 @default.
- W2969121442 cites W2480880131 @default.
- W2969121442 cites W2605135824 @default.
- W2969121442 cites W2626637010 @default.
- W2969121442 cites W2756196406 @default.
- W2969121442 cites W2788388592 @default.
- W2969121442 cites W2810602713 @default.
- W2969121442 cites W2904616874 @default.
- W2969121442 cites W3093287223 @default.
- W2969121442 cites W2770298516 @default.
- W2969121442 hasPublicationYear "2019" @default.
- W2969121442 type Work @default.
- W2969121442 sameAs 2969121442 @default.
- W2969121442 citedByCount "0" @default.
- W2969121442 crossrefType "posted-content" @default.
- W2969121442 hasAuthorship W2969121442A5062895972 @default.
- W2969121442 hasAuthorship W2969121442A5064045506 @default.
- W2969121442 hasAuthorship W2969121442A5079570262 @default.
- W2969121442 hasAuthorship W2969121442A5091771063 @default.
- W2969121442 hasConcept C105339364 @default.
- W2969121442 hasConcept C111919701 @default.
- W2969121442 hasConcept C11413529 @default.
- W2969121442 hasConcept C115903868 @default.
- W2969121442 hasConcept C119857082 @default.
- W2969121442 hasConcept C120314980 @default.
- W2969121442 hasConcept C12713177 @default.
- W2969121442 hasConcept C127413603 @default.
- W2969121442 hasConcept C143587482 @default.
- W2969121442 hasConcept C146978453 @default.
- W2969121442 hasConcept C154945302 @default.
- W2969121442 hasConcept C17744445 @default.
- W2969121442 hasConcept C199539241 @default.
- W2969121442 hasConcept C204323151 @default.
- W2969121442 hasConcept C2776359362 @default.
- W2969121442 hasConcept C39890363 @default.
- W2969121442 hasConcept C41008148 @default.
- W2969121442 hasConcept C48103436 @default.
- W2969121442 hasConcept C50644808 @default.
- W2969121442 hasConcept C774472 @default.
- W2969121442 hasConcept C94625758 @default.
- W2969121442 hasConcept C97541855 @default.
- W2969121442 hasConcept C98045186 @default.
- W2969121442 hasConceptScore W2969121442C105339364 @default.
- W2969121442 hasConceptScore W2969121442C111919701 @default.
- W2969121442 hasConceptScore W2969121442C11413529 @default.
- W2969121442 hasConceptScore W2969121442C115903868 @default.
- W2969121442 hasConceptScore W2969121442C119857082 @default.
- W2969121442 hasConceptScore W2969121442C120314980 @default.
- W2969121442 hasConceptScore W2969121442C12713177 @default.
- W2969121442 hasConceptScore W2969121442C127413603 @default.
- W2969121442 hasConceptScore W2969121442C143587482 @default.
- W2969121442 hasConceptScore W2969121442C146978453 @default.
- W2969121442 hasConceptScore W2969121442C154945302 @default.
- W2969121442 hasConceptScore W2969121442C17744445 @default.
- W2969121442 hasConceptScore W2969121442C199539241 @default.
- W2969121442 hasConceptScore W2969121442C204323151 @default.
- W2969121442 hasConceptScore W2969121442C2776359362 @default.
- W2969121442 hasConceptScore W2969121442C39890363 @default.
- W2969121442 hasConceptScore W2969121442C41008148 @default.
- W2969121442 hasConceptScore W2969121442C48103436 @default.
- W2969121442 hasConceptScore W2969121442C50644808 @default.
- W2969121442 hasConceptScore W2969121442C774472 @default.
- W2969121442 hasConceptScore W2969121442C94625758 @default.
- W2969121442 hasConceptScore W2969121442C97541855 @default.
- W2969121442 hasConceptScore W2969121442C98045186 @default.
- W2969121442 hasLocation W29691214421 @default.
- W2969121442 hasOpenAccess W2969121442 @default.
- W2969121442 hasPrimaryLocation W29691214421 @default.
- W2969121442 hasRelatedWork W1546568226 @default.
- W2969121442 hasRelatedWork W1601125311 @default.
- W2969121442 hasRelatedWork W1983015815 @default.
- W2969121442 hasRelatedWork W2025448855 @default.
- W2969121442 hasRelatedWork W2111770102 @default.
- W2969121442 hasRelatedWork W2153427071 @default.
- W2969121442 hasRelatedWork W2225266074 @default.
- W2969121442 hasRelatedWork W2272929109 @default.
- W2969121442 hasRelatedWork W2770887358 @default.
- W2969121442 hasRelatedWork W2791803308 @default.
- W2969121442 hasRelatedWork W2980297462 @default.
- W2969121442 hasRelatedWork W2998093422 @default.
- W2969121442 hasRelatedWork W3015849478 @default.
- W2969121442 hasRelatedWork W3123636359 @default.