The αμ Search Algorithm for the Game of Bridge
This work provides a novel method for Bridge players and AI researchers, though it appears incremental as it builds on existing techniques for incomplete information games.
The authors tackled the problem of incomplete information games by developing the αμ search algorithm, which assumes perfect information for opponents and addresses strategy fusion and non-locality issues, applying it to Bridge.
αμ is an anytime heuristic search algorithm for incomplete information games that assumes perfect information for the opponents. αμ addresses the strategy fusion and non-locality problems encountered by Perfect Information Monte Carlo sampling. In this paper αμ is applied to the game of Bridge.