AI learns to play flappy bird using neuro-evolution, implemented in Rust using macroquad

Similar flappy bird design to the previous repo, but this time, a non supervised genetic algorithm is used to train a neural network to play the game

This repository contains the code for a Flappy Bird game with NEAT AI. NEAT is a genetic algorithm that can be used to train neural networks to play games. In this project, a NEAT neural network is trained to play Flappy Bird. The neural network learns to control the bird so that it can avoid the pipes and score points.

An gentic algorithm based agent which can learn to play Flappy Bird Game.

FlappyBirdAI utilizes NEAT Python to train an AI to master Flappy Bird. Employing genetic algorithms, this project evolves neural networks over generations to improve gameplay performance automatically. Ideal for developers and AI enthusiasts exploring game AI and evolutionary algorithms in Python.

Flappy bird implements a DQN agent in Unity, with Python handling the reinforcement learning logic. The agent gets observation states by interacting with the Unity environment.

The classic flappy bird game made using Python and PyGame