RLSnake is a project where we implement and train artificial intelligence to play the classic Snake game using reinforcement learning. We leverage advanced deep learning techniques to enable our artificial intelligence to effectively navigate the dynamic and changing Snake game environment.
The game environment is custom-built using PyGame, a popular library for creating 2D games in Python. Our environment provides the necessary framework for the agent to interact with the game world, receive feedback, and learn from its actions. The environment is compatible with reinforcement learning libraries, making it suitable for training AI agents.
The project uses the Deep Q-Learning (DQN): A popular RL algorithm that uses a neural network to approximate the Q-value function.
