A Novel Point-based Algorithm for Multi-agent Control Using the Common Information Approach
This work provides an incremental improvement for researchers in multi-agent systems by enhancing computational efficiency in solving control problems.
The paper tackled the challenge of solving multi-agent stochastic control problems by addressing the large action space in the coordinator's POMDP derived from the Common Information approach, resulting in a new algorithm called CHSVI that optimally solves benchmark problems.
The Common Information (CI) approach provides a systematic way to transform a multi-agent stochastic control problem to a single-agent partially observed Markov decision problem (POMDP) called the coordinator's POMDP. However, such a POMDP can be hard to solve due to its extraordinarily large action space. We propose a new algorithm for multi-agent stochastic control problems, called coordinator's heuristic search value iteration (CHSVI), that combines the CI approach and point-based POMDP algorithms for large action spaces. We demonstrate the algorithm through optimally solving several benchmark problems.