Learning strongly deterministic even linear languages from positive examples

Takeshi Koshiba; Erkki Mäkinen; Yuji Takada

Learning strongly deterministic even linear languages from positive examples

Takeshi Koshiba, Erkki Mäkinen, Yuji Takada

研究成果: Conference contribution

抄録

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].

本文言語	English
ホスト出版物のタイトル	Algorithmic Learning Theory - 6th International Workshop, ALT 1995, Proceedings
出版社	Springer Verlag
ページ	41-54
ページ数	14
巻	997
ISBN（印刷版）	3540604545, 9783540604549
出版ステータス	Published - 1995
外部発表	はい
イベント	6th International Workshop on Algorithmic Learning Theory, ALT 1995 - Fukuoka, Japan 継続期間: 1995 10月 18 → 1995 10月 20

出版物シリーズ

名前	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
巻	997
ISSN（印刷版）	0302-9743
ISSN（電子版）	1611-3349

Other

Other	6th International Workshop on Algorithmic Learning Theory, ALT 1995
国/地域	Japan
City	Fukuoka
Period	95/10/18 → 95/10/20

ASJC Scopus subject areas

理論的コンピュータサイエンス
コンピュータサイエンス（全般）

他のファイルとリンク

引用スタイル

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).

研究成果: Conference 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",

note = "6th International Workshop on Algorithmic Learning Theory, ALT 1995 ; Conference date: 18-10-1995 Through 20-10-1995",

}

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

T2 - 6th International Workshop on Algorithmic Learning Theory, ALT 1995

Y2 - 18 October 1995 through 20 October 1995

ER -

Learning strongly deterministic even linear languages from positive examples

抄録

出版物シリーズ

Other

ASJC Scopus subject areas

他のファイルとリンク

フィンガープリント

引用スタイル