DeepRL

My implementations of Deep Reinforcement Learning algorithms DDPG and NAF tested on OpenAI gym environments Hopper and LunarLander. DDPG works well, but NAF still has some implementation issues.

TODO

• Fix NAF
• Try different environments