Refining Domain Theories Expressed as Finite-State Automata

Richard Maclin; Jude W. Shavlik

doi:10.1016/B978-1-55860-200-7.50107-0

Refining Domain Theories Expressed as Finite-State Automata

Richard Maclin, Jude W. Shavlik

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

8 Scopus citations

Abstract

The kbann system uses neural networks to refine domain theories. Currently, domain knowledge in kbann is expressed as nonrecursive, propositional rules. We extend kbann to domain theories expressed as finite-state automata. We apply finite-state KBANN to the task of predicting how proteins fold, producing a small but statistically significant gain in accuracy over both a standard neural network approach and a non-learning algorithm from the biological literature. Our method shows promise at solving this and other real-world problems that can be described in terms of statedependent decisions.

Original language	English (US)
Title of host publication	Proceedings of the 8th International Workshop on Machine Learning, ICML 1991
Editors	Lawrence A. Birnbaum, Gregg C. Collins
Publisher	Morgan Kaufmann Publishers, Inc.
Pages	524-528
Number of pages	5
ISBN (Electronic)	1558602003, 9781558602007
DOIs	https://doi.org/10.1016/B978-1-55860-200-7.50107-0
State	Published - 1991
Externally published	Yes
Event	8th International Workshop on Machine Learning, ICML 1991 - Evanston, United States Duration: Jun 1 1991 → …

Publication series

Name	Proceedings of the 8th International Workshop on Machine Learning, ICML 1991

Conference

Conference	8th International Workshop on Machine Learning, ICML 1991
Country/Territory	United States
City	Evanston
Period	6/1/91 → …

Bibliographical note

Publisher Copyright:
© ICML 1989.All rights reserved

Publisher link

10.1016/B978-1-55860-200-7.50107-0

Find It

Find It at U of M Libraries

Cite this

Maclin, R., & Shavlik, J. W. (1991). Refining Domain Theories Expressed as Finite-State Automata. In L. A. Birnbaum, & G. C. Collins (Eds.), Proceedings of the 8th International Workshop on Machine Learning, ICML 1991 (pp. 524-528). (Proceedings of the 8th International Workshop on Machine Learning, ICML 1991). Morgan Kaufmann Publishers, Inc.. https://doi.org/10.1016/B978-1-55860-200-7.50107-0

Refining Domain Theories Expressed as Finite-State Automata. / Maclin, Richard; Shavlik, Jude W.
Proceedings of the 8th International Workshop on Machine Learning, ICML 1991. ed. / Lawrence A. Birnbaum; Gregg C. Collins. Morgan Kaufmann Publishers, Inc., 1991. p. 524-528 (Proceedings of the 8th International Workshop on Machine Learning, ICML 1991).

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

Maclin, R & Shavlik, JW 1991, Refining Domain Theories Expressed as Finite-State Automata. in LA Birnbaum & GC Collins (eds), Proceedings of the 8th International Workshop on Machine Learning, ICML 1991. Proceedings of the 8th International Workshop on Machine Learning, ICML 1991, Morgan Kaufmann Publishers, Inc., pp. 524-528, 8th International Workshop on Machine Learning, ICML 1991, Evanston, United States, 6/1/91. https://doi.org/10.1016/B978-1-55860-200-7.50107-0

@inproceedings{f1b40a4688a942a3b58a95e6f7a02eb2,

title = "Refining Domain Theories Expressed as Finite-State Automata",

abstract = "The kbann system uses neural networks to refine domain theories. Currently, domain knowledge in kbann is expressed as nonrecursive, propositional rules. We extend kbann to domain theories expressed as finite-state automata. We apply finite-state KBANN to the task of predicting how proteins fold, producing a small but statistically significant gain in accuracy over both a standard neural network approach and a non-learning algorithm from the biological literature. Our method shows promise at solving this and other real-world problems that can be described in terms of statedependent decisions.",

author = "Richard Maclin and Shavlik, {Jude W.}",

year = "1991",

doi = "10.1016/B978-1-55860-200-7.50107-0",

language = "English (US)",

series = "Proceedings of the 8th International Workshop on Machine Learning, ICML 1991",

publisher = "Morgan Kaufmann Publishers, Inc.",

pages = "524--528",

editor = "Birnbaum, {Lawrence A.} and Collins, {Gregg C.}",

booktitle = "Proceedings of the 8th International Workshop on Machine Learning, ICML 1991",

address = "United States",

}

TY - GEN

T1 - Refining Domain Theories Expressed as Finite-State Automata

AU - Maclin, Richard

AU - Shavlik, Jude W.

PY - 1991

Y1 - 1991

N2 - The kbann system uses neural networks to refine domain theories. Currently, domain knowledge in kbann is expressed as nonrecursive, propositional rules. We extend kbann to domain theories expressed as finite-state automata. We apply finite-state KBANN to the task of predicting how proteins fold, producing a small but statistically significant gain in accuracy over both a standard neural network approach and a non-learning algorithm from the biological literature. Our method shows promise at solving this and other real-world problems that can be described in terms of statedependent decisions.

AB - The kbann system uses neural networks to refine domain theories. Currently, domain knowledge in kbann is expressed as nonrecursive, propositional rules. We extend kbann to domain theories expressed as finite-state automata. We apply finite-state KBANN to the task of predicting how proteins fold, producing a small but statistically significant gain in accuracy over both a standard neural network approach and a non-learning algorithm from the biological literature. Our method shows promise at solving this and other real-world problems that can be described in terms of statedependent decisions.

UR - http://www.scopus.com/inward/record.url?scp=33747943914&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33747943914&partnerID=8YFLogxK

U2 - 10.1016/B978-1-55860-200-7.50107-0

DO - 10.1016/B978-1-55860-200-7.50107-0

M3 - Conference contribution

AN - SCOPUS:33747943914

T3 - Proceedings of the 8th International Workshop on Machine Learning, ICML 1991

SP - 524

EP - 528

BT - Proceedings of the 8th International Workshop on Machine Learning, ICML 1991

A2 - Birnbaum, Lawrence A.

A2 - Collins, Gregg C.

PB - Morgan Kaufmann Publishers, Inc.

T2 - 8th International Workshop on Machine Learning, ICML 1991

Y2 - 1 June 1991

ER -

Refining Domain Theories Expressed as Finite-State Automata

Abstract

Publication series

Conference

Bibliographical note

Publisher link

Find It

Other files and links

Fingerprint

Cite this