Train Your Reinforcement Learning Agents at the OpenAI Gym

Originally published at: https://developer.nvidia.com/blog/train-reinforcement-learning-agents-openai-gym/

Today OpenAI, a non-profit artificial intelligence research company, launched OpenAI Gym, a toolkit for developing and comparing reinforcement learning algorithms. It supports teaching agents everything from walking to playing games like Pong or Go. John Schulman is a researcher at OpenAI. OpenAI researcher John Schulman shared some details about his organization, and how OpenAI Gym will make it easier for AI researchers to…

Thanks Mark for such a great blog on RL. I think RL will be the coming DL hotspot. For openai gym, does it support existing DL frameworks? such as Caffe, Torch, Tensorflow, Theano, etc..?

Yes, OpenAI Gym doesn't make any restrictions on how you write your algorithms, so you can use any of these frameworks.

Does it have a ROS interface?

http://erlerobotics.com/doc...
Here you will find a toolkit extending the OpenAI Gym for robotics, which supports ROS and Gazebo.