This is a tic tac toe in which the computer AI is referred as MENACE which is an implementation of Donald Michie algorithms. This is a classic proof by Donald about showing that learning machine is possible. It start with random moves and learns through out the game. For learning the menace takes random moves.In initial games each move has equal points but as it plays game it rewards points to the move taken in certain game position based on what is the final result of that corresponding game. -1/+1/+3 reward points for loss/tie/win game. The current limit of number of game for learning is 300 games.
