Robot Learning under Misspecification
confidence-aware robot learning from human input
To enable robots to use human input as guidance on desired behaviors, system designers typically equip them with a representation of possible objectives the person could care about. However, these designers and, hence, the techniques for robot learning that they employ operate on the assumption that the human's desired objective can always be captured by the robot's representation. In our work, we investigate what the robot can do when this assumption breaks. We propose a method where the robot reasons explicitly about how well it can explain human inputs given its hypothesis space and use that situational confidence to inform how it should incorporate human input.
Project materials:
- Code for confidence-aware learning from physical human corrections and demonstrations on a Jaco 7DOF robotic manipulator.
- Video showcasing our experimental results in a 12-person user study.
- Talk from the Conference on Robot Learning (CoRL) 2018.
- Poster illustrating our method and results.
Andreea Bobu, Andrea Bajcsy , Jaime F. Fisac, andAnca D. Dragan. Learning under Misspecified Objective Spaces. Conference on Robot Learning (CoRL), 2018.Andreea Bobu, Andrea Bajcsy , Jaime F. Fisac, Sampada Deglurkar, andAnca D. Dragan. Quantifying Hypothesis Space Misspecification in Learning from Human-Robot Demonstrations and Physical Corrections. IEEE Transactions on Robotics (T-RO), 2019.Andreea Bobu. Detecting Hypothesis Space Misspecification in Robot Learning from Human Input. Human-Robot Interaction Pioneers Workshop 2020.