liuyandong1988/Maze_ReinforcementLearning

The maze instance by different RL approaches

Maze instance by different reinforcement learning approaches

1. Environment

• python = 3.6.0
• gym = 0.15.4
• tensorflow-gpu = 1.13.1
• keras = 2.2.4
• gym-gridword

2. RL algorithm

• Three cheeses by q learning
• Maze by dqn(https://github.com/liuyandong1988/Maze_ReinforcementLearning/tree/master/maze_dqn)
• sarsa one and more step(https://github.com/liuyandong1988/Maze_ReinforcementLearning/tree/master/maze_sarsa)
• spring_2020