This project is an implementation of a Connect Four AI using a minimax algorithm with alpha-beta pruning. The game playable in a representation in the console and its a head to head game against the AI.
The AI is built using the minimax algorithm with alpha-beta pruning. The psudeo code for the algorithm can be found here. For the evaluation function we used simple score given condition such as three in a row and two in a row. The state of the game is represented using a numpy matrix and certain helper functions is used to update the state.
- A better heurstic evaulation function.
- Some performance optimizations to make the algorithm run faster.
- AI against AI games.
- Grahpical Interface.