Matches in SemOpenAlex for { <https://semopenalex.org/work/W2891963971> ?p ?o ?g. }
- W2891963971 abstract "A variety of cooperative multi-agent control problems require agents to achieve individual goals while contributing to collective success. This multi-goal multi-agent setting poses difficulties for recent algorithms, which primarily target settings with a single global reward, due to two new challenges: efficient exploration for learning both individual goal attainment and cooperation for others' success, and credit-assignment for interactions between actions and goals of different agents. To address both challenges, we restructure the problem into a novel two-stage curriculum, in which single-agent goal attainment is learned prior to learning multi-agent cooperation, and we derive a new multi-goal multi-agent policy gradient with a credit function for localized credit assignment. We use a function augmentation scheme to bridge value and policy functions across the curriculum. The complete architecture, called CM3, learns significantly faster than direct adaptations of existing algorithms on three challenging multi-goal multi-agent problems: cooperative navigation in difficult formations, negotiating multi-vehicle lane changes in the SUMO traffic simulator, and strategic cooperation in a Checkers environment." @default.
- W2891963971 created "2018-09-27" @default.
- W2891963971 creator A5046703129 @default.
- W2891963971 creator A5050227531 @default.
- W2891963971 creator A5063634505 @default.
- W2891963971 creator A5066144350 @default.
- W2891963971 creator A5071231555 @default.
- W2891963971 date "2018-09-13" @default.
- W2891963971 modified "2023-09-27" @default.
- W2891963971 title "CM3: Cooperative Multi-goal Multi-stage Multi-agent Reinforcement Learning" @default.
- W2891963971 cites W1504212531 @default.
- W2891963971 cites W1536258751 @default.
- W2891963971 cites W1542941925 @default.
- W2891963971 cites W2012812921 @default.
- W2891963971 cites W2021702776 @default.
- W2891963971 cites W2097381042 @default.
- W2891963971 cites W2103561211 @default.
- W2891963971 cites W2107544712 @default.
- W2891963971 cites W2109910161 @default.
- W2891963971 cites W2120846115 @default.
- W2891963971 cites W2122253967 @default.
- W2891963971 cites W2122763142 @default.
- W2891963971 cites W2145339207 @default.
- W2891963971 cites W2147492008 @default.
- W2891963971 cites W2155027007 @default.
- W2891963971 cites W2165150801 @default.
- W2891963971 cites W2165698076 @default.
- W2891963971 cites W2296073425 @default.
- W2891963971 cites W2580495915 @default.
- W2891963971 cites W2594829461 @default.
- W2891963971 cites W2604873668 @default.
- W2891963971 cites W2768629321 @default.
- W2891963971 cites W2807741983 @default.
- W2891963971 cites W2891663723 @default.
- W2891963971 cites W2892013712 @default.
- W2891963971 cites W2894141523 @default.
- W2891963971 cites W2895865957 @default.
- W2891963971 cites W2903709398 @default.
- W2891963971 cites W2904455790 @default.
- W2891963971 cites W2911616846 @default.
- W2891963971 cites W2946606218 @default.
- W2891963971 cites W2962764167 @default.
- W2891963971 cites W2962938168 @default.
- W2891963971 cites W2963407617 @default.
- W2891963971 cites W2963625099 @default.
- W2891963971 cites W2963637944 @default.
- W2891963971 cites W2963650250 @default.
- W2891963971 cites W2963658727 @default.
- W2891963971 cites W2963747324 @default.
- W2891963971 cites W2963864421 @default.
- W2891963971 cites W2963881016 @default.
- W2891963971 cites W2964014087 @default.
- W2891963971 cites W2964338167 @default.
- W2891963971 cites W2970879379 @default.
- W2891963971 cites W2975185915 @default.
- W2891963971 cites W3093287223 @default.
- W2891963971 cites W567721252 @default.
- W2891963971 cites W2426267443 @default.
- W2891963971 hasPublicationYear "2018" @default.
- W2891963971 type Work @default.
- W2891963971 sameAs 2891963971 @default.
- W2891963971 citedByCount "2" @default.
- W2891963971 countsByYear W28919639712019 @default.
- W2891963971 countsByYear W28919639712021 @default.
- W2891963971 crossrefType "posted-content" @default.
- W2891963971 hasAuthorship W2891963971A5046703129 @default.
- W2891963971 hasAuthorship W2891963971A5050227531 @default.
- W2891963971 hasAuthorship W2891963971A5063634505 @default.
- W2891963971 hasAuthorship W2891963971A5066144350 @default.
- W2891963971 hasAuthorship W2891963971A5071231555 @default.
- W2891963971 hasConcept C100776233 @default.
- W2891963971 hasConcept C10138342 @default.
- W2891963971 hasConcept C126322002 @default.
- W2891963971 hasConcept C134306372 @default.
- W2891963971 hasConcept C136197465 @default.
- W2891963971 hasConcept C14036430 @default.
- W2891963971 hasConcept C144133560 @default.
- W2891963971 hasConcept C154945302 @default.
- W2891963971 hasConcept C17744445 @default.
- W2891963971 hasConcept C199539241 @default.
- W2891963971 hasConcept C199776023 @default.
- W2891963971 hasConcept C2775924081 @default.
- W2891963971 hasConcept C33923547 @default.
- W2891963971 hasConcept C41008148 @default.
- W2891963971 hasConcept C45237549 @default.
- W2891963971 hasConcept C71924100 @default.
- W2891963971 hasConcept C77618280 @default.
- W2891963971 hasConcept C78458016 @default.
- W2891963971 hasConcept C86803240 @default.
- W2891963971 hasConcept C97541855 @default.
- W2891963971 hasConceptScore W2891963971C100776233 @default.
- W2891963971 hasConceptScore W2891963971C10138342 @default.
- W2891963971 hasConceptScore W2891963971C126322002 @default.
- W2891963971 hasConceptScore W2891963971C134306372 @default.
- W2891963971 hasConceptScore W2891963971C136197465 @default.
- W2891963971 hasConceptScore W2891963971C14036430 @default.
- W2891963971 hasConceptScore W2891963971C144133560 @default.
- W2891963971 hasConceptScore W2891963971C154945302 @default.
- W2891963971 hasConceptScore W2891963971C17744445 @default.
- W2891963971 hasConceptScore W2891963971C199539241 @default.