Advanced

Learning to evaluate conditional partial plans

Nowaczyk, Sławomir and Malec, Jacek LU (2007) Sixth International Conference on Machine Learning and Applications, 2007. ICMLA 2007. In [Host publication title missing] p.235-240
Abstract
We study agents situated in partially observable environments, who do not have sufficient resources to create conformant plans. Instead, they generate plans which are conditional and partial, execute or simulate them, and learn to evaluate their quality from experience. Our agent employs an incomplete symbolic deduction system based on active logic and situation calculus for reasoning about actions and their consequences. An inductive logic programming algorithm generalises observations and deduced knowledge, allowing the agent to execute a good plan. We show results of using PROGOL learning algorithm to distinguish "bad" plans early in the reasoning process, before too many resources are wasted on considering them. We show that additional... (More)
We study agents situated in partially observable environments, who do not have sufficient resources to create conformant plans. Instead, they generate plans which are conditional and partial, execute or simulate them, and learn to evaluate their quality from experience. Our agent employs an incomplete symbolic deduction system based on active logic and situation calculus for reasoning about actions and their consequences. An inductive logic programming algorithm generalises observations and deduced knowledge, allowing the agent to execute a good plan. We show results of using PROGOL learning algorithm to distinguish "bad" plans early in the reasoning process, before too many resources are wasted on considering them. We show that additional knowledge needs to be provided before learning can be successful, but argue that the benefits achieved make it worthwhile. Finally, we identify several assumptions made by PROGOL, shared by other similarly universal algorithms, which are well justified in general, but fail to exploit the properties of the class of problems faced by rational agents. (Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
in
[Host publication title missing]
pages
6 pages
publisher
IEEE--Institute of Electrical and Electronics Engineers Inc.
conference name
Sixth International Conference on Machine Learning and Applications, 2007. ICMLA 2007.
external identifiers
  • Scopus:47349104416
ISBN
978-0-7695-3069-7
DOI
10.1109/ICMLA.2007.101
language
English
LU publication?
yes
id
1a685f4f-7d68-43f6-b34d-0baa60f7a8f5 (old id 4679164)
date added to LUP
2014-09-25 13:05:39
date last changed
2016-10-13 04:39:29
@misc{1a685f4f-7d68-43f6-b34d-0baa60f7a8f5,
  abstract     = {We study agents situated in partially observable environments, who do not have sufficient resources to create conformant plans. Instead, they generate plans which are conditional and partial, execute or simulate them, and learn to evaluate their quality from experience. Our agent employs an incomplete symbolic deduction system based on active logic and situation calculus for reasoning about actions and their consequences. An inductive logic programming algorithm generalises observations and deduced knowledge, allowing the agent to execute a good plan. We show results of using PROGOL learning algorithm to distinguish "bad" plans early in the reasoning process, before too many resources are wasted on considering them. We show that additional knowledge needs to be provided before learning can be successful, but argue that the benefits achieved make it worthwhile. Finally, we identify several assumptions made by PROGOL, shared by other similarly universal algorithms, which are well justified in general, but fail to exploit the properties of the class of problems faced by rational agents.},
  author       = {Nowaczyk, Sławomir and Malec, Jacek},
  isbn         = {978-0-7695-3069-7},
  language     = {eng},
  pages        = {235--240},
  publisher    = {ARRAY(0xb60ddb8)},
  series       = {[Host publication title missing]},
  title        = {Learning to evaluate conditional partial plans},
  url          = {http://dx.doi.org/10.1109/ICMLA.2007.101},
  year         = {2007},
}