Pong NEAT is an implementation of the classic arcade game Pong with a twist: the paddle is controlled not by a human, but by an AI agent trained via NeuroEvolution of Augmenting Topologies (NEAT). NEAT is a genetic algorithm that evolves artificial neural networks with the ability to learn and adapt in complex environments.
- Classic Pong gameplay with a modern AI opponent.
- NEAT implemented for real-time learning and adaptation.
- Ability to train the AI over multiple sessions to improve performance.
- Visualization of neural network performance and decision-making process.
You don't! You watch and see if the AI learns over time to become a Pong champion. However, you can reset the environment and tweak the AI settings to see how different parameters affect learning.
- Python 3.8+
- Pygame
- NEAT-Python
Clone the repository and navigate to the directory!