Using Deep Q Networks for Atari Breakout

Keras implementation of deep-q-network for Atari breakout of OpenAI Gym.

Implements Deep Q-network (DQN) in Keras following the architecture proposed in the 2013 paper by V. Mnih et al., "Playing Atari with Deep Reinforcement Learning": arXiv:1312.5602. See: http://www.davidqiu.com:8888/research/nature14236.pdf