Implementation of the Deep Q-learning algorithm to solve MDPs

Julia implementation of DeepMind's Deep Q Learning algorithm described in "Playing Atari with Deep Reinforcement Learning"

Unfinished project that combines DeepQLearning.jl with Mill.jl in order to solve POMDPs with nonparametric distributional states (variable number of input states creating distribution)