More than 100 girl geeks met at OpenAI HQ in San Francisco to to connect with OpenAI researchers and learn about their recent work in reinforcement learning, robotics, AI policy, and more!

In this talk, I'll introduce how the Robotics team uses reinforcement learning (RL) to learn dexterous in-hand manipulation policies which can perform vision-based object reorientation on a physical Shadow Dexterous Hand. Our policies are trained in a simulated environment in which we randomize many physical properties of the system and an object’s appearance – the method is known as Domain Randomization. Although the training happens entirely in a simulator, our policies transfer to the physical robot well.