An implementation of OpenAI's Slitherin' Request for Research. This is a multiplayer snake game intended for reinforcement learning implemented in python with pygame. Parts of this code are borrowed from https://github.com/sds-dubois/snake.ai to implement the game agent.
- Q-Learning
- Policy Gradient
- Baselines (Greedy, Opportunist, Weighted Hill-Climbing)
- Deep Q-Learning
- Minimax
- Monte Carlo
- Conversion to OpenAI gym environment