Approximation Schemes for Stochastic Teams with Weakly Coupled Agents

Abstract

The main objective of this Papers is to develop a new iterative approach towards the Solution of dynamic stochastic teams when the coupling among a subset of agents is weak. either throughth the state dynamics or through the performance index. In each case. the weak coupling is characterized in terms of number of small (pertuhrltion) parameters. When these parameter values are set equal to Zero . the original fairly complex dynamic team. with a quasiclassical or a non classica information pattern is decomposed into relatively simpler storhastic control and team problems.the solution of which make up t he zerorh order approximation (in a function space) to the team-optimal solution of the original problem. Using this asastarling point for an ite ra tion. We show that approximations of all orders can be obtained by solving a sequence of stochastic control and/or simplper team problems. The paper studies. particlliarly two agents LQG team.With quasiciassicai or non classical information patterns. but the approach is applicable to other models (with more than two agents and non-Gaussian sta tislics) as well.