Teamtheory analyzes cooperative decision-makingamongmultiple agents that share acommon objective while possessing different information.In team problems with nonclassical information structures,the classical separatio...Teamtheory analyzes cooperative decision-makingamongmultiple agents that share acommon objective while possessing different information.In team problems with nonclassical information structures,the classical separation collapses and prevents the formulation of a classical dynamic programming(DP)decomposition.This article focuses on a two-member dynamic team.Weshow how to construct information states that are independent of control strategies for both a centralised manager and individual members,which allows us to formulate a classical DP over the space of the decisions,recovering a separation-like architecture in decentralised settings.展开更多
基金supported in part by National Science Foundation NSF under Grants CNS-2401007,CMMI-2348381,IIS-2415478in part by MathWorks.
文摘Teamtheory analyzes cooperative decision-makingamongmultiple agents that share acommon objective while possessing different information.In team problems with nonclassical information structures,the classical separation collapses and prevents the formulation of a classical dynamic programming(DP)decomposition.This article focuses on a two-member dynamic team.Weshow how to construct information states that are independent of control strategies for both a centralised manager and individual members,which allows us to formulate a classical DP over the space of the decisions,recovering a separation-like architecture in decentralised settings.
