Train Your Reinforcement Learning Agents at the OpenAI Gym

openai learning network reinforcement gym agents train robot nvidia atari playing devblogs simulated architectures schulman et al control game used