RL-Pytorch-Cartpole

Reinforcement Learning tutorial by pytorch

Implemented algorithms:

• Deep Q-Learning (DQN)
• Double DQN
• Dueling DQN
• Policy Gradient

Learning & Playing

Max step : 1000

DQN

Playing :

Double DQN

Playing :

Dueling DQN

Playing :

Policy Gradient

More stable, Faster(not needed replay memory), more simple(not needed customizing policy)

Playing :