An Investigation of Minimax Search for Evolving Ayo/Awari Player

Minimax search is the most basic class of algorithms for game playing, but when use in isolation, it fails for a certain class of games called Mancala that Ayo, which is also known as Awari or Awale belongs. However, when unswervingly complemented with a powerful refinement procedure, it provides an efficient scheme for a move generation. This paper investigates some heuristics based on hybrid of minimax and refinement, to evolve an Ayo player. The technique has advantage that a search is astutely conducted in a shallow horizon game tree. The evolved player was tested against Awale shareware and appealing results that show considerable improvement over those obtained from evolutionary techniques were obtained.