This project implements a Java program that evolves feed-forward neural networks to control two classic arcade games: Breakout and Pacman. The goal is to optimize the performance of the neural networks using a genetic algorithm.
- Utilizes feed-forward neural networks for game control.
- Implements a genetic algorithm to optimize network weights.
- Provides APIs for interacting with Breakout and Pacman games.
- Generates populations of neural networks and evaluates their performance using fitness functions.
- Applies evolutionary process to improve network performance over multiple generations.
- Java Development Kit (JDK) installed on your system.
- Eclipse IDE or any other Java IDE for development.
- Clone this repository to your local machine:
git clone https://github.com/your-username/your-repository.git
-
Open the project in your Java IDE.
-
Start exploring and modifying the code according to your requirements.
-
Run the
Main
class to start the program. -
Follow the instructions provided in the console to interact with the games and observe the evolution of neural networks.