In this project, we will implement AI players in the board game Reversi (also called Othello) with the utilization of Tree data structure.
Due to the 8x8 size of the grid, there are many potential moves that can be made each turn, which provides some unique challenges when implementing decision-making algorithms. We want to compare the performance of different decision-making algorithms.
We will focus on two algorithms in the project:
-
Minimax: This is an algorithm widely used in two player turn-based game. We would implement various versions of the AI player with this algorithm based on different strategies presented by different value evaluation functions
- The greedy strategy: Aiming for the maximum number of pieces of its side after a certain depth of the game tree
- The positional strategy: Aiming for occupying certain position on the board to gain positional advantage. For instance, corner pieces are immune to being flipped.
- Mobility strategy: Aiming for as many possible moves as possible
-
Monte Carlo Tree Search (MCTS): An heuristic search algorithm for decision making used in complex engines like AlphaGo
These instructions will get you a copy of the project up and running on your local machine for development and testing purposes.
The project is built on python 3.9. Please install python 3.9 with tkinter installation
Please install all python libraries in requirement.txt.
- Haoze Deng
- Peifeng Zhang
- Man Chon Ho
- Alexander Nicholas Conway