Pong AI - Pong Game with Q-Learning

This repository contains a Pong game where two agents compete against each other using Q-Learning, a Reinforcement Learning algorithm.
The agents learn autonomously, improving their gameplay strategies over time by aiming to hit the ball and score points.

Features

• 🕹️ Pong game with real-time visualization using pygame.
• 🤖 Q-Learning-based AI: Two agents (Player 1 and Player 2) train themselves to play the game.
• 📊 Real-time score display: The current score is displayed at the top of the screen during gameplay.
• 🎯 Custom reward system: Rewards are provided for different agent behaviors, such as collisions, perfect bounces, and errors.

Installation

1. Clone the repository:

2. git clone https://github.com/NumberZeroo/PongAI.git
   cd PongAI

3. Create a virtual environment (optional but recommended):

4. python3 -m venv venv
   source venv/bin/activate  # Windows: venv\Scripts\activate

Usage

To start the training and run the game:

python main.py

Customization

Modify parameters in main.py to tweak training settings, game speed, and AI behavior.

License

This project is licensed under the MIT LICENCE.