Developed a Tic Tac Toe game featuring an opponent powered by a machine learning algorithm, specifically Q-learning, a type of reinforcement learning. This project demonstrates the practical application of machine learning in game development, providing an interactive experience with a learning-driven opponent.
- Machine Learning Opponent: Utilizes Q-learning to adapt and improve strategies over time.
- User-Friendly Interface: Built with Pygame for an engaging graphical user experience.
- Performance Metrics: Achieves a win rate of over 80% against inexperienced players and a tie rate of approximately 90% against optimal human gameplay.
- Clone the Repository:
git clone https://github.com/JustDud/AI-Tic-Tac-Toe.git
- Navigate to the Project Directory:
cd AI-Tic-Tac-Toe - Install Dependencies:
Ensure you have Python installed. Then, install the required packages:
pip install -r requirements.txt
Run the game using the following command:
python main.pyFollow the on-screen instructions to play against the machine learning-powered opponent.
Comprehensive documentation is available, detailing development stages, bug fixes, evaluations, stakeholder feedback, and analysis. View the Documentation
- Programming Language: Python
- Libraries: Pygame, NumPy, Matplotlib
- Machine Learning: Q-learning algorithm
The following image provides a snapshot of the Tic Tac Toe game in action, including the graphical user interface and a performance graph illustrating the Q-learning algorithm's progress.
Contributions are welcome! Please fork the repository and submit a pull request with your proposed changes. Ensure adherence to the project's coding standards and include relevant tests.
For inquiries or feedback, feel free to connect with me on LinkedIn.
Keywords: Machine Learning Tic Tac Toe, Q-learning, Reinforcement Learning Game, Python Pygame, Machine Learning Game Development