Asking Easy Questions: A User-Friendly Approach to Active Reward
Learning
release_53wiv6e5rjd2lpb2gqwk6alomi
by
Erdem Bıyık, Malayandi Palan, Nicholas C. Landolfi, Dylan P.
Losey, Dorsa Sadigh
2019
Abstract
Robots can learn the right reward function by querying a human expert.
Existing approaches attempt to choose questions where the robot is most
uncertain about the human's response; however, they do not consider how easy it
will be for the human to answer! In this paper we explore an information gain
formulation for optimally selecting questions that naturally account for the
human's ability to answer. Our approach identifies questions that optimize the
trade-off between robot and human uncertainty, and determines when these
questions become redundant or costly. Simulations and a user study show our
method not only produces easy questions, but also ultimately results in faster
reward learning.
In text/plain
format
Archived Files and Locations
application/pdf 8.6 MB
file_npvlrquqrjb7dicrxxowkvw3e4
|
arxiv.org (repository) web.archive.org (webarchive) |
1910.04365v1
access all versions, variants, and formats of this works (eg, pre-prints)