0 votes
1 view
in AI and Deep Learning by (17.6k points)
I am making a little Java game, based on the game Blokus.

I'm just a beginner in Java, and I plan to implement advanced artificial intelligence concepts. I already have a random AI (picks a random valid move) and an AI with a simple move-rating mechanism. I now want an AI which should be as good as possible.

My question is: Which AI-concept would be suitable for my purpose? The minimax algorithm seems to be a valid choice, but how do I adapt it to a 4-player-game? Are there better concepts for a game like Blokus?

1 Answer

0 votes
by (73k points)

Min-max is difficult to implement for a 4 player game because:

  • The decision tree grows exponentially, so you face issues regarding memory and/or computation time (log(medMoves)=N steps). For a 4 player game, this will be N/4. If N is 8 then you will have only 2 moves ahead for each player.
  • Player collusion is hard to deal with. In a realistic game, some players may help each other out even though they are not playing for the same team. This will cause them to deviate from their personal 'maximum'.

If you want Minmax, you have to do a lot of pruning to make it feasible. I suggest learning a few patterns so that AI would know how to respond. This can be done using a neural network or reinforcement learning using just a few tweaks.

If you want to learn Artificial Intelligencethen check out this AI Course by Intellipaat.

Welcome to Intellipaat Community. Get your technical queries answered by top developers !