skeptical_lurker comments on “Human-level control through deep reinforcement learning”—computer learns 49 different games

[skeptical_lurker]

Ok thanks, finally got that working.