The altruistic robot : Do what i want, not just what i say
(2017) 11th International Conference on Scalable Uncertainty Management, SUM 2017 In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) 10564 LNAI. p.149-162- Abstract
As autonomous robots expand their application beyond research labs and production lines, they must work in more flexible and less well defined environments. To escape the requirement for exhaustive instruction and stipulated preference ordering, a robot’s operation must involve choices between alternative actions, guided by goals. We describe a robot that learns these goals from humans by considering the timeliness and context of instructions and rewards as evidence of the contours and gradients of an unknown human utility function. In turn, this underlies a choice-theory based rational preference relationship. We examine how the timing of requests, and contexts in which they arise, can lead to actions that pre-empt requests using... (More)
As autonomous robots expand their application beyond research labs and production lines, they must work in more flexible and less well defined environments. To escape the requirement for exhaustive instruction and stipulated preference ordering, a robot’s operation must involve choices between alternative actions, guided by goals. We describe a robot that learns these goals from humans by considering the timeliness and context of instructions and rewards as evidence of the contours and gradients of an unknown human utility function. In turn, this underlies a choice-theory based rational preference relationship. We examine how the timing of requests, and contexts in which they arise, can lead to actions that pre-empt requests using methods we term contemporaneous entropy learning and context sensitive learning. We provide experiments on these two methods to demonstrate their usefulness in guiding a robot’s actions.
(Less)
- author
- Billingsley, Richard ; Billingsley, John ; Gärdenfors, Peter LU ; Peppas, Pavlos ; Prade, Henri ; Skillicorn, David and Williams, Mary Anne
- publishing date
- 2017-01-01
- type
- Chapter in Book/Report/Conference proceeding
- publication status
- published
- subject
- host publication
- Scalable Uncertainty Management - 11th International Conference, SUM 2017. Proceedings
- series title
- Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
- editor
- Moral, Serafin ; Sanchez, Daniel ; Marin, Nicolas and Pivert, Olivier
- volume
- 10564 LNAI
- pages
- 149 - 162
- publisher
- Springer
- conference name
- 11th International Conference on Scalable Uncertainty Management, SUM 2017
- conference location
- Granada, Spain
- conference dates
- 2017-10-04 - 2017-10-06
- external identifiers
-
- scopus:85030864522
- ISSN
- 0302-9743
- 1611-3349
- ISBN
- 978-3-319-67582-4
- 9783319675817
- DOI
- 10.1007/978-3-319-67582-4_11
- language
- English
- LU publication?
- no
- id
- fd5545f5-07ea-45ac-b465-d6286e5106a0
- date added to LUP
- 2019-06-12 16:37:28
- date last changed
- 2024-04-30 12:33:37
@inproceedings{fd5545f5-07ea-45ac-b465-d6286e5106a0,
abstract = {{<p>As autonomous robots expand their application beyond research labs and production lines, they must work in more flexible and less well defined environments. To escape the requirement for exhaustive instruction and stipulated preference ordering, a robot’s operation must involve choices between alternative actions, guided by goals. We describe a robot that learns these goals from humans by considering the timeliness and context of instructions and rewards as evidence of the contours and gradients of an unknown human utility function. In turn, this underlies a choice-theory based rational preference relationship. We examine how the timing of requests, and contexts in which they arise, can lead to actions that pre-empt requests using methods we term contemporaneous entropy learning and context sensitive learning. We provide experiments on these two methods to demonstrate their usefulness in guiding a robot’s actions.</p>}},
author = {{Billingsley, Richard and Billingsley, John and Gärdenfors, Peter and Peppas, Pavlos and Prade, Henri and Skillicorn, David and Williams, Mary Anne}},
booktitle = {{Scalable Uncertainty Management - 11th International Conference, SUM 2017. Proceedings}},
editor = {{Moral, Serafin and Sanchez, Daniel and Marin, Nicolas and Pivert, Olivier}},
isbn = {{978-3-319-67582-4}},
issn = {{0302-9743}},
language = {{eng}},
month = {{01}},
pages = {{149--162}},
publisher = {{Springer}},
series = {{Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)}},
title = {{The altruistic robot : Do what i want, not just what i say}},
url = {{http://dx.doi.org/10.1007/978-3-319-67582-4_11}},
doi = {{10.1007/978-3-319-67582-4_11}},
volume = {{10564 LNAI}},
year = {{2017}},
}