For learning purposes, I am trying to implement the minimax algorithm for the ColorShapeLinks game, which is similar to connect 4, except the fact that it combines both shape and color as the winning conditions, with a shape having priority over color. A color is associated with only one shape, so there can only be X blue or O red.

I have previously applied the minimax to TicTacToe, but I think that's a lot simpler than this one.

My question is: which heuristic function could be used to evaluate the states for this game?

I'm thinking of checking each window of the board, where the new piece is put, and then compare the differences of shapes and colors and calculate the total of each predetermined value before (all heuristic). However, I think that's a little bit too simple, right?

Also, is there any chance that the AI can be made with Local Search algorithms, like Hill Climbing, Annealing, or GA? I think we can't, since the state configurations are not complete, right? If there are any additional reasons for this, please kindly guide me through, I am pretty new in this research :D



You must log in to answer this question.