robotics
The Robot Asks About the Part You Failed to Teach
A preference learner identifies undersupervised reward features, targets its questions there and tells the person what to notice.
Summary
A preference learner identifies undersupervised reward features, targets its questions there and tells the person what to notice.
Human teaching is selective: a demonstration can make some task features clear while leaving others poorly specified. This method uses the same uncertainty that identifies weakly supervised features to choose targeted preference questions and explain which features deserve attention. In simulation, the informed prior beat the standard active-learning baseline by as much as 30.8 percent when only a few questions were allowed. In a user study, both targeted queries and explanations improved reward recovery and reinforced one another. The paper does not establish universal human factors or deployment safety. It shows how a robot can carry more of the burden when earlier teaching left a gap.
Why it matters
A preference learner identifies undersupervised reward features, targets its questions there and tells the person what to notice.
Limits and context
- In simulation, the informed prior beat the standard active-learning baseline by as much as 30.8 percent when only a few questions were allowed.
- The paper does not establish universal human factors or deployment safety.
Key claims
A preference learner identifies undersupervised reward features, targets its questions there and tells the person what to notice.
Qualification: In simulation, the informed prior beat the standard active-learning baseline by as much as 30.8 percent when only a few questions were allowed.
Evidence: source-2026-10-16-016
Sources
- arXiv preprint 2610.22285, original v1arXiv / submitting research authors · primary research
Corrections
No corrections have been recorded for this story.