Solving Transition-Independent Multi-agent MDPs with Sparse Interactions (Extended version)
This addresses coordination challenges in multi-agent systems for domains like robotics or logistics, but it is incremental as it builds on existing transition-independent MDP frameworks.
The paper tackles the problem of finding optimal joint policies in cooperative multi-agent MDPs with sparse interactions, proposing a new solver called CoRe that uses conditional return graphs for efficient computation, resulting in reduced runtime and solving previously unsolvable problems.
In cooperative multi-agent sequential decision making under uncertainty, agents must coordinate to find an optimal joint policy that maximises joint value. Typical algorithms exploit additive structure in the value function, but in the fully-observable multi-agent MDP setting (MMDP) such structure is not present. We propose a new optimal solver for transition-independent MMDPs, in which agents can only affect their own state but their reward depends on joint transitions. We represent these dependencies compactly in conditional return graphs (CRGs). Using CRGs the value of a joint policy and the bounds on partially specified joint policies can be efficiently computed. We propose CoRe, a novel branch-and-bound policy search algorithm building on CRGs. CoRe typically requires less runtime than the available alternatives and finds solutions to problems previously unsolvable.