linyiLYi/snake-ai
A deep reinforcement learning project that trains neural network agents to play the classic Snake game.
Not currently ranked — collecting fresh signals.
star history
This project implements AI agents trained via deep reinforcement learning to play the Snake game automatically. It includes two model architectures: a Multi-Layer Perceptron (MLP) agent and a Convolution Neural Network (CNN) agent, with the CNN version achieving higher average scores. Training uses Stable-Baselines3 and PyTorch, with progress tracked via Tensorboard. Pre-trained model weights are provided for testing.
Frequently asked
- What is linyiLYi/snake-ai?
- A deep reinforcement learning project that trains neural network agents to play the classic Snake game.
- Is snake-ai open source?
- Yes — linyiLYi/snake-ai is open source, released under the Apache-2.0 license.
- What language is snake-ai written in?
- linyiLYi/snake-ai is primarily written in Python.
- How popular is snake-ai?
- linyiLYi/snake-ai has 1.8k stars on GitHub.
- Where can I find snake-ai?
- linyiLYi/snake-ai is on GitHub at https://github.com/linyiLYi/snake-ai.