Learning strongly deterministic even linear languages from positive examples

Takeshi Koshiba, Erkki Mäkinen, Yuji Takada

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

We consider the problem of learning deterministic even lin­ear languages from positive examples. By a “deterministic” even linear language we mean a language generated by an LR(k) even linear gram­mar. We introduce a natural subclass of LR(k) even linear languages, called LR(k) in the strong sense, and show that this subclass is learn- able in the limit from positive examples. Furthermore, we propose a learning algorithm that identifies this subclass in the limit with almost linear time in updating conjectures. As a corollary, in terms of even Un­ear grammars, we have a learning algorithm for k-reversible languages that is more efficient than the one proposed by Angluin[Ang82].

Original languageEnglish
Title of host publicationAlgorithmic Learning Theory - 6th International Workshop, ALT 1995, Proceedings
PublisherSpringer Verlag
Pages41-54
Number of pages14
Volume997
ISBN (Print)3540604545, 9783540604549
Publication statusPublished - 1995
Externally publishedYes
Event6th International Workshop on Algorithmic Learning Theory, ALT 1995 - Fukuoka, Japan
Duration: 1995 Oct 181995 Oct 20

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume997
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other6th International Workshop on Algorithmic Learning Theory, ALT 1995
CountryJapan
CityFukuoka
Period95/10/1895/10/20

Fingerprint

Learning algorithms
Grammar
Learning Algorithm
Updating
Linear Time
Corollary
Learning
Language

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Koshiba, T., Mäkinen, E., & Takada, Y. (1995). Learning strongly deterministic even linear languages from positive examples. In Algorithmic Learning Theory - 6th International Workshop, ALT 1995, Proceedings (Vol. 997, pp. 41-54). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 997). Springer Verlag.

Learning strongly deterministic even linear languages from positive examples. / Koshiba, Takeshi; Mäkinen, Erkki; Takada, Yuji.

Algorithmic Learning Theory - 6th International Workshop, ALT 1995, Proceedings. Vol. 997 Springer Verlag, 1995. p. 41-54 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 997).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Koshiba, T, Mäkinen, E & Takada, Y 1995, Learning strongly deterministic even linear languages from positive examples. in Algorithmic Learning Theory - 6th International Workshop, ALT 1995, Proceedings. vol. 997, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 997, Springer Verlag, pp. 41-54, 6th International Workshop on Algorithmic Learning Theory, ALT 1995, Fukuoka, Japan, 95/10/18.
Koshiba T, Mäkinen E, Takada Y. Learning strongly deterministic even linear languages from positive examples. In Algorithmic Learning Theory - 6th International Workshop, ALT 1995, Proceedings. Vol. 997. Springer Verlag. 1995. p. 41-54. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
Koshiba, Takeshi ; Mäkinen, Erkki ; Takada, Yuji. / Learning strongly deterministic even linear languages from positive examples. Algorithmic Learning Theory - 6th International Workshop, ALT 1995, Proceedings. Vol. 997 Springer Verlag, 1995. pp. 41-54 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{65cff30e86274904b0282392087d8eaa,
title = "Learning strongly deterministic even linear languages from positive examples",
abstract = "We consider the problem of learning deterministic even lin­ear languages from positive examples. By a “deterministic” even linear language we mean a language generated by an LR(k) even linear gram­mar. We introduce a natural subclass of LR(k) even linear languages, called LR(k) in the strong sense, and show that this subclass is learn- able in the limit from positive examples. Furthermore, we propose a learning algorithm that identifies this subclass in the limit with almost linear time in updating conjectures. As a corollary, in terms of even Un­ear grammars, we have a learning algorithm for k-reversible languages that is more efficient than the one proposed by Angluin[Ang82].",
author = "Takeshi Koshiba and Erkki M{\"a}kinen and Yuji Takada",
year = "1995",
language = "English",
isbn = "3540604545",
volume = "997",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "41--54",
booktitle = "Algorithmic Learning Theory - 6th International Workshop, ALT 1995, Proceedings",
address = "Germany",

}

TY - GEN

T1 - Learning strongly deterministic even linear languages from positive examples

AU - Koshiba, Takeshi

AU - Mäkinen, Erkki

AU - Takada, Yuji

PY - 1995

Y1 - 1995

N2 - We consider the problem of learning deterministic even lin­ear languages from positive examples. By a “deterministic” even linear language we mean a language generated by an LR(k) even linear gram­mar. We introduce a natural subclass of LR(k) even linear languages, called LR(k) in the strong sense, and show that this subclass is learn- able in the limit from positive examples. Furthermore, we propose a learning algorithm that identifies this subclass in the limit with almost linear time in updating conjectures. As a corollary, in terms of even Un­ear grammars, we have a learning algorithm for k-reversible languages that is more efficient than the one proposed by Angluin[Ang82].

AB - We consider the problem of learning deterministic even lin­ear languages from positive examples. By a “deterministic” even linear language we mean a language generated by an LR(k) even linear gram­mar. We introduce a natural subclass of LR(k) even linear languages, called LR(k) in the strong sense, and show that this subclass is learn- able in the limit from positive examples. Furthermore, we propose a learning algorithm that identifies this subclass in the limit with almost linear time in updating conjectures. As a corollary, in terms of even Un­ear grammars, we have a learning algorithm for k-reversible languages that is more efficient than the one proposed by Angluin[Ang82].

UR - http://www.scopus.com/inward/record.url?scp=84947938312&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84947938312&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84947938312

SN - 3540604545

SN - 9783540604549

VL - 997

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 41

EP - 54

BT - Algorithmic Learning Theory - 6th International Workshop, ALT 1995, Proceedings

PB - Springer Verlag

ER -