gitlele92/rl-algorithms-pytorch

Reinforcement Algorithms - Policy Gradient, Q Learning, Double Q Learning, Deep Q Learning and Double Deep Q Learning

RL-Algorithms-Pytorch

Reinforcement Algorithms - Policy Gradient, Q Learning, Double Q Learning, Deep Q Learning and Double Deep Q Learning

Results

• DQN obtained the average score of 17/21
• DDQN obtained the average score of 19/21

Both of them are playing like a pro player. Check out the videos below.

DQN : https://youtu.be/tplpUiHNxPU

DDQN : https://youtu.be/aK_Wrgg5sIM

DDQN is really playing like pro player.