1/ Curious how the @OpenAI team runs a massive machine learning project – like training a robot hand to manipulate physical objects e.g. the Rubik's cube?
In this report, @atpaino offers a window into how they design their hypotheses, run their experiments & analyze the results.
We've trained an AI system to solve the Rubik's Cube with a human-like robot hand.
This is an unprecedented level of dexterity for a robot, and is hard even for humans to do.
The system trains in an imperfect simulation and quickly adapts to reality: https://t.co/O04izt3KvO