XOShift AI Agent

Introduction

This project implements an AI agent to play the board game XOShift using Python.
The agent leverages a minimax algorithm with alpha-beta pruning, iterative deepening, and a heuristic evaluation function to make optimal moves within a set time limit.
The repository also includes the XOShift game logic, a performance evaluation script, and a simple UI.

Table of Contents

• Introduction
• Game Preview
• Features
• Installation
• Usage
• Project Structure
• License

Game Preview

Figure: The XOShift game environment with the AI agent making moves.
_{Note: The GIF playback speed has been increased for better loading performance.}

Features

• Minimax search with alpha-beta pruning.
• Iterative deepening with time limit control.
• Heuristic-based board evaluation (mobility, threats, forks, open twos).
• Performance evaluation via evaluate_performance.py.
• Complete XOShift rules implemented in game.py.

Installation

1. Clone the repository:

   git clone https://github.com/azizi-zahra/xoshift-ai-agent.git
   cd xoshift-ai-agent

2. Install dependencies:

   pip install -r requirements.txt

3. Ensure you are using Python 3.10+.

Usage

• Run the game with UI:

  python main.py

• Run performance evaluation:

  python evaluate_performance.py

• Adjust agent settings (time limit, depth) in your_agent.py under the CONFIGURATION section.

Project Structure

• game.py → XOShift game logic and rules.

• your_agent.py → AI agent implementation (minimax + heuristics).

• evaluate_performance.py → Script to test agent performance.

• ui.py → Simple game interface.

• agent_loader.py, agent_utils.py, utils.py → Supporting utilities.

• assets/ → Fonts and other resources.

License

This project is licensed under the MIT License.