Paper Accepted into HRI22: Correct Me If I'm Wrong: Using Non-Experts to Repair Reinforcement Learning Policies
An overview of the failure corrections we consider in this work.
Our work on correcting robot policies using non-experts is accepted to HRI 2022. We will present this work during the conference in March 2022. In this collaboration with Chris Pek, we ask people who do not necessarily have programming and robotics experience to correct high-level robot behavior after the robot fails in a deployment environment. More info will follow.