Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313334471> ?p ?o ?g. }
- W4313334471 endingPage "514" @default.
- W4313334471 startingPage "483" @default.
- W4313334471 abstract "In artificial multi-agent systems, the ability to learn collaborative policies is predicated upon the agents' communication skills: they must be able to encode the information received from the environment and learn how to share it with other agents as required by the task at hand. We present a deep reinforcement learning approach, Connectivity Driven Communication (CDC), that facilitates the emergence of multi-agent collaborative behaviour only through experience. The agents are modelled as nodes of a weighted graph whose state-dependent edges encode pair-wise messages that can be exchanged. We introduce a graph-dependent attention mechanisms that controls how the agents' incoming messages are weighted. This mechanism takes into full account the current state of the system as represented by the graph, and builds upon a diffusion process that captures how the information flows on the graph. The graph topology is not assumed to be known a priori, but depends dynamically on the agents' observations, and is learnt concurrently with the attention mechanism and policy in an end-to-end fashion. Our empirical results show that CDC is able to learn effective collaborative policies and can over-perform competing learning algorithms on cooperative navigation tasks." @default.
- W4313334471 created "2023-01-06" @default.
- W4313334471 creator A5010581004 @default.
- W4313334471 creator A5059716182 @default.
- W4313334471 date "2022-12-29" @default.
- W4313334471 modified "2023-09-26" @default.
- W4313334471 title "Learning multi-agent coordination through connectivity-driven communication" @default.
- W4313334471 cites W1498447134 @default.
- W4313334471 cites W1508056422 @default.
- W4313334471 cites W1512522810 @default.
- W4313334471 cites W1518858799 @default.
- W4313334471 cites W1542941925 @default.
- W4313334471 cites W1579184372 @default.
- W4313334471 cites W1591723504 @default.
- W4313334471 cites W1639167632 @default.
- W4313334471 cites W1813318478 @default.
- W4313334471 cites W1964883026 @default.
- W4313334471 cites W1990647940 @default.
- W4313334471 cites W1997881026 @default.
- W4313334471 cites W2016273060 @default.
- W4313334471 cites W2024649868 @default.
- W4313334471 cites W2034692547 @default.
- W4313334471 cites W2049561033 @default.
- W4313334471 cites W2061214064 @default.
- W4313334471 cites W2064675550 @default.
- W4313334471 cites W2076063813 @default.
- W4313334471 cites W2099618002 @default.
- W4313334471 cites W2106348760 @default.
- W4313334471 cites W2119120935 @default.
- W4313334471 cites W2127800791 @default.
- W4313334471 cites W2128453677 @default.
- W4313334471 cites W2132600297 @default.
- W4313334471 cites W2132848998 @default.
- W4313334471 cites W2136398072 @default.
- W4313334471 cites W2137813581 @default.
- W4313334471 cites W2145339207 @default.
- W4313334471 cites W2147492008 @default.
- W4313334471 cites W2153073482 @default.
- W4313334471 cites W2165321212 @default.
- W4313334471 cites W2166715144 @default.
- W4313334471 cites W2173200640 @default.
- W4313334471 cites W2257979135 @default.
- W4313334471 cites W2292533394 @default.
- W4313334471 cites W2311061258 @default.
- W4313334471 cites W2496794451 @default.
- W4313334471 cites W2586491110 @default.
- W4313334471 cites W2602275733 @default.
- W4313334471 cites W2617547828 @default.
- W4313334471 cites W2758442112 @default.
- W4313334471 cites W2804610937 @default.
- W4313334471 cites W285701945 @default.
- W4313334471 cites W2908261578 @default.
- W4313334471 cites W2919115771 @default.
- W4313334471 cites W2962764167 @default.
- W4313334471 cites W2963998174 @default.
- W4313334471 cites W2966859223 @default.
- W4313334471 cites W2981038142 @default.
- W4313334471 cites W2982316857 @default.
- W4313334471 cites W2991653934 @default.
- W4313334471 cites W2998367975 @default.
- W4313334471 cites W3035096461 @default.
- W4313334471 cites W3035098634 @default.
- W4313334471 cites W3042612721 @default.
- W4313334471 cites W3106014076 @default.
- W4313334471 cites W3136999308 @default.
- W4313334471 cites W3192813613 @default.
- W4313334471 cites W4206560593 @default.
- W4313334471 cites W4214717370 @default.
- W4313334471 cites W4252284432 @default.
- W4313334471 cites W613360614 @default.
- W4313334471 doi "https://doi.org/10.1007/s10994-022-06286-6" @default.
- W4313334471 hasPublicationYear "2022" @default.
- W4313334471 type Work @default.
- W4313334471 citedByCount "1" @default.
- W4313334471 countsByYear W43133344712023 @default.
- W4313334471 crossrefType "journal-article" @default.
- W4313334471 hasAuthorship W4313334471A5010581004 @default.
- W4313334471 hasAuthorship W4313334471A5059716182 @default.
- W4313334471 hasBestOaLocation W43133344711 @default.
- W4313334471 hasConcept C104317684 @default.
- W4313334471 hasConcept C111472728 @default.
- W4313334471 hasConcept C120314980 @default.
- W4313334471 hasConcept C132525143 @default.
- W4313334471 hasConcept C138885662 @default.
- W4313334471 hasConcept C154945302 @default.
- W4313334471 hasConcept C165696696 @default.
- W4313334471 hasConcept C185592680 @default.
- W4313334471 hasConcept C38652104 @default.
- W4313334471 hasConcept C41008148 @default.
- W4313334471 hasConcept C41550386 @default.
- W4313334471 hasConcept C55493867 @default.
- W4313334471 hasConcept C66746571 @default.
- W4313334471 hasConcept C75553542 @default.
- W4313334471 hasConcept C80444323 @default.
- W4313334471 hasConcept C97541855 @default.
- W4313334471 hasConceptScore W4313334471C104317684 @default.
- W4313334471 hasConceptScore W4313334471C111472728 @default.
- W4313334471 hasConceptScore W4313334471C120314980 @default.