Python RL

This repository serves as a home for my work and experiments on reinforcement learning algos.

Lunar Lander

The ppo lunar lander is my own tensorflow implemetation of the PPO algorithm.

- https://github.com/nikhilbarhate99/PPO-PyTorch/blob/master/PPO.py
- https://github.com/DavidCastilloAlvarado/PPO_reinforcement_learning/blob/master/PPO_pendulum.py
- https://blog.varunajayasiri.com/ml/ppo_pytorch.html
- (paper) https://arxiv.org/abs/1707.06347

Snake

Created my own environment for snake [gym-snake] as a gym registered environment. Solved with stable baselines.

See the Readme in SnakeRL folder.

NEAT

Neat is an evolutionary algorithm that evolves neural networks.

(nothing here yet will upload in future)