CUMIN hosted its second paper reading group of Michaelmas on the paper titled “Solving Rubik’s Cube with a Robot Hand”. Presented by CUMIN President Henry Pulver, the session gave a brief overview and discussion of OpenAI’s recent paper, which was published just a couple of weeks prior. The abstract of the paper is presented below.

In case you missed this session, you can check out their paper and website here. Follow our Facebook event for our up-to-date lineup of papers for this term, and like our Facebook page to keep informed of the other events that we’ll be doing!

CUMIN’s paper reading group happens weekly on Wednesdays 2pm in the North Room, Department of Engineering. No prior reading of the paper is required, just show up to find out more. All are welcome!

Check out our paper reading page for information on other regular paper reading groups that are going on around Cambridge.


We demonstrate that models trained only in simulation can be used to solve a manipulation problem of unprecedented complexity on a real robot. This is made possible by two key components: a novel algorithm, which we call automatic domain randomization (ADR) and a robot platform built for machine learning. ADR automatically generates a distribution over randomized environments of ever-increasing difficulty. Control policies and vision state estimators trained with ADR exhibit vastly improved sim2real transfer. For control policies, memory-augmented models trained on an ADR-generated distribution of environments show clear signs of emergent meta-learning at test time. The combination of ADR with our custom robot platform allows us to solve a Rubik’s cube with a humanoid robot hand, which involves both control and state estimation problems. Videos summarizing our results are available:


Please enter your comment!
Please enter your name here