The Utility of EBL in Recursive Domain Theories

Devika Subramanian; Ronen Feldman

The Utility of EBL in Recursive Domain Theories

Devika Subramanian, Ronen Feldman

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

18 Scopus citations

Abstract

We investigate the utility of explanation-based learning in recursive domain theories and examine the cost of using macro-rules in these theories. The compilation options in a recursive domain theory range from constructing partial unwindings of the recursive rules to converting recursive rules into iterative ones. We compare these options against using appropriately ordered rules in the original domain theory and demonstrate that unless we make very strong assumptions about the nature of the distribution of future problems, it is not profitable to form recursive macro-rules via explanation-based learning in these domains.

Original language	English
Title of host publication	Proceedings of the 8th National Conference on Artificial Intelligence, AAAI 1990
Publisher	AAAI Press
Pages	942-949
Number of pages	8
ISBN (Electronic)	026251057X, 9780262510578
State	Published - 1990
Externally published	Yes
Event	8th National Conference on Artificial Intelligence, AAAI 1990 - Boston, United States Duration: 29 Jul 1990 → 3 Aug 1990

Publication series

Name	Proceedings of the 8th National Conference on Artificial Intelligence, AAAI 1990

Conference

Conference	8th National Conference on Artificial Intelligence, AAAI 1990
Country/Territory	United States
City	Boston
Period	29/07/90 → 3/08/90

Bibliographical note

Publisher Copyright:
Copyright © 1990, AAAI (www.aaai.org). All rights reserved.

Cite this

@inproceedings{7634e6c0744a4f75bff8e45104465644,

title = "The Utility of EBL in Recursive Domain Theories",

abstract = "We investigate the utility of explanation-based learning in recursive domain theories and examine the cost of using macro-rules in these theories. The compilation options in a recursive domain theory range from constructing partial unwindings of the recursive rules to converting recursive rules into iterative ones. We compare these options against using appropriately ordered rules in the original domain theory and demonstrate that unless we make very strong assumptions about the nature of the distribution of future problems, it is not profitable to form recursive macro-rules via explanation-based learning in these domains.",

author = "Devika Subramanian and Ronen Feldman",

note = "Publisher Copyright: Copyright {\textcopyright} 1990, AAAI (www.aaai.org). All rights reserved.; 8th National Conference on Artificial Intelligence, AAAI 1990 ; Conference date: 29-07-1990 Through 03-08-1990",

year = "1990",

language = "אנגלית",

series = "Proceedings of the 8th National Conference on Artificial Intelligence, AAAI 1990",

publisher = "AAAI Press",

pages = "942--949",

booktitle = "Proceedings of the 8th National Conference on Artificial Intelligence, AAAI 1990",

}

Subramanian, D & Feldman, R 1990, The Utility of EBL in Recursive Domain Theories. in Proceedings of the 8th National Conference on Artificial Intelligence, AAAI 1990. Proceedings of the 8th National Conference on Artificial Intelligence, AAAI 1990, AAAI Press, pp. 942-949, 8th National Conference on Artificial Intelligence, AAAI 1990, Boston, United States, 29/07/90.

The Utility of EBL in Recursive Domain Theories. / Subramanian, Devika; Feldman, Ronen.
Proceedings of the 8th National Conference on Artificial Intelligence, AAAI 1990. AAAI Press, 1990. p. 942-949 (Proceedings of the 8th National Conference on Artificial Intelligence, AAAI 1990).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - The Utility of EBL in Recursive Domain Theories

AU - Subramanian, Devika

AU - Feldman, Ronen

PY - 1990

Y1 - 1990

N2 - We investigate the utility of explanation-based learning in recursive domain theories and examine the cost of using macro-rules in these theories. The compilation options in a recursive domain theory range from constructing partial unwindings of the recursive rules to converting recursive rules into iterative ones. We compare these options against using appropriately ordered rules in the original domain theory and demonstrate that unless we make very strong assumptions about the nature of the distribution of future problems, it is not profitable to form recursive macro-rules via explanation-based learning in these domains.

AB - We investigate the utility of explanation-based learning in recursive domain theories and examine the cost of using macro-rules in these theories. The compilation options in a recursive domain theory range from constructing partial unwindings of the recursive rules to converting recursive rules into iterative ones. We compare these options against using appropriately ordered rules in the original domain theory and demonstrate that unless we make very strong assumptions about the nature of the distribution of future problems, it is not profitable to form recursive macro-rules via explanation-based learning in these domains.

UR - http://www.scopus.com/inward/record.url?scp=0343300298&partnerID=8YFLogxK

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:0343300298

T3 - Proceedings of the 8th National Conference on Artificial Intelligence, AAAI 1990

SP - 942

EP - 949

BT - Proceedings of the 8th National Conference on Artificial Intelligence, AAAI 1990

PB - AAAI Press

T2 - 8th National Conference on Artificial Intelligence, AAAI 1990

Y2 - 29 July 1990 through 3 August 1990

ER -

The Utility of EBL in Recursive Domain Theories

Abstract

Publication series

Conference

Bibliographical note

Other files and links

Fingerprint

Cite this