Asking Smarter Questions: Targeted Active Preference Learning for Underspecified Rewards

Paper recorded by Signals 4 on 2026-10-15 in cs.LG. Abstract reproduced from arXiv; link to the original below.

Published 2026-10-15 on arXiv · recorded by Signals 4 on 2026-10-16

Category: cs.LG · 机器学习 · first seen 2026-10-16

Abstract

Reward learning typically assumes the human's input covers everything that matters for the task. In practice, humans can only attend to so much at once, so their teaching naturally captures some features---or aspects of the task---better than others. Prior work has shown how the robot can identify which features were poorly supervised directly from its teaching data. But the current solution is to

Read on arXiv →