We consider decentralized control of Markov decision processes and give complexity bounds on the worst-case running time for algorithms that find optimal solutions. Generalizations of both the fully observable case and the partially observable case that allow for decentralized control are described. For even two agents, the finite-horizon problems corresponding to both of these models are hard for nondeterministic exponential time. These complexity results illustrate a fundamental difference between centralized and decentralized control of Markov decision processes. In contrast to the problems involving centralized control, the problems we consider provably do not admit polynomial-time algorithms. Furthermore, assuming EXP = NEXP, the problems require superexponential time to solve in the worst case.

POMDPs and their decentralized multiagent counterparts, DEC-POMDPs, offer a rich framework for sequential decision making under uncertainty. Their high computational complexity, however, presents an important research challenge. One way to address the intractable memory requirements of current algorithms is based on representing agent policies as finite-state controllers. Using this representation, we propose a new approach that formulates the problem as a nonlinear program, which defines an optimal policy of a desired size for each agent. This new formulation allows a wide range of powerful nonlinear programming algorithms to be used to solve POMDPs and DEC-POMDPs. Although solving the NLP optimally is often intractable, the results we obtain using an off-the-shelf optimization method are competitive with state-of-the-art POMDP algorithms and outperform state-of-the-art DEC-POMDP algorithms. Our approach is easy to implement and it opens up promising research directions for solving POMDPs and DEC-POMDPs using nonlinear programming methods.

Coordination of distributed agents is required for problems arising in many areas, including multi-robot systems, networking and e-commerce. As a formal framework for such problems, we use the decentralized partially observable Markov decision process (DEC-POMDP). Though much work has been done on optimal dynamic programming algorithms for the single-agent version of the problem, optimal algorithms for the multiagent case have been elusive. The main contribution of this paper is an optimal policy iteration algorithm for solving DEC-POMDPs. The algorithm uses stochastic finite-state controllers to represent policies. The solution can include a correlation device, which allows agents to correlate their actions without communicating. This approach alternates between expanding the controller and performing value-preserving transformations, which modify the controller without sacrificing value. We present two efficient value-preserving transformations: one can reduce the size of the controller and the other can improve its value while keeping the size fixed. Empirical results demonstrate the usefulness of value-preserving transformations in increasing value while keeping controller size to a minimum. To broaden the applicability of the approach, we also present a heuristic version of the policy iteration algorithm, which sacrifices convergence to optimality. This algorithm further reduces the size of the controllers at each step by assuming that probability distributions over the other agents' actions are known. While this assumption may not hold in general, it helps produce higher quality solutions in our test problems.

No abstract

Developing scalable algorithms for solving partially observable Markov decision processes (POMDPs) is an important challenge. One approach that effectively addresses the intractable memory requirements of POMDP algorithms is based on representing POMDP policies as finitestate controllers. In this paper, we illustrate some fundamental disadvantages of existing techniques that use controllers. We then propose a new approach that formulates the problem as a quadratically constrained linear program (QCLP), which defines an optimal controller of a desired size. This representation allows a wide range of powerful nonlinear programming algorithms to be used to solve POMDPs. Although QCLP optimization techniques guarantee only local optimality, the results we obtain using an existing optimization method show significant solution improvement over the state-of-the-art techniques. The results open up promising research directions for solving large POMDPs using nonlinear programming methods.

No abstract

No abstract

No abstract