README.md

neural_rl

Neural Reinforcement Learning

Uses Arcade Learning Environment, RL Glue, and pylearn2

This is an attempt to recreate an experiment by DeepMind: http://arxiv.org/abs/1312.5602