Deep reinforcement algorithms

Implementations of classic Deep RL algorithms, mostly based on CS 294: Deep Reinforcement Learning at Berekley

Current state:

• Policy Gradient:

  Basically my solution to heavily refactored CS294 second homework. Tensorflow ლ(ಠ益ಠლ)
  

• DQN

  Still based on CS294 but implemented in  PyTorch ✧ﾟ･: *ヽ(◕ヮ◕ヽ)
  

  • Discrete action space
  • Continous action space

Solved classic control environments: