Learning deterministic even linear languages from positive examples

Takeshi Koshiba, Erkki Mäkinen, Yuji Takada

Research output: Contribution to journalArticle

19 Citations (Scopus)

Abstract

We consider the problem of learning deterministic even linear languages from positive examples. We show that, for any nonnegative integer k, the class of LR(k) even linear languages is not learnable from positive examples while there is a subclass called LRS(k), which is a natural subclass of LR(k) in the strong sense, learnable from positive examples. Our learning algorithm identifies this subclass in the limit with almost linear time in updating conjectures. As a corollary, in terms of even linear grammars, we have a learning algorithm for k-reversible languages that is more efficient than the one proposed by Angluin.

Original languageEnglish
Pages (from-to)63-79
Number of pages17
JournalTheoretical Computer Science
Volume185
Issue number1
DOIs
Publication statusPublished - 1997 Oct 10
Externally publishedYes

Fingerprint

Learning algorithms
Learning Algorithm
Grammar
Updating
Linear Time
Corollary
Non-negative
Integer
Learning
Language
Class

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

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

In: Theoretical Computer Science, Vol. 185, No. 1, 10.10.1997, p. 63-79.

Research output: Contribution to journalArticle

Koshiba, Takeshi ; Mäkinen, Erkki ; Takada, Yuji. / Learning deterministic even linear languages from positive examples. In: Theoretical Computer Science. 1997 ; Vol. 185, No. 1. pp. 63-79.
@article{e74a95205a314128aec52dbb57436a4e,
title = "Learning deterministic even linear languages from positive examples",
abstract = "We consider the problem of learning deterministic even linear languages from positive examples. We show that, for any nonnegative integer k, the class of LR(k) even linear languages is not learnable from positive examples while there is a subclass called LRS(k), which is a natural subclass of LR(k) in the strong sense, learnable from positive examples. Our learning algorithm identifies this subclass in the limit with almost linear time in updating conjectures. As a corollary, in terms of even linear grammars, we have a learning algorithm for k-reversible languages that is more efficient than the one proposed by Angluin.",
author = "Takeshi Koshiba and Erkki M{\"a}kinen and Yuji Takada",
year = "1997",
month = "10",
day = "10",
doi = "10.1016/S0304-3975(97)00016-9",
language = "English",
volume = "185",
pages = "63--79",
journal = "Theoretical Computer Science",
issn = "0304-3975",
publisher = "Elsevier",
number = "1",

}

TY - JOUR

T1 - Learning deterministic even linear languages from positive examples

AU - Koshiba, Takeshi

AU - Mäkinen, Erkki

AU - Takada, Yuji

PY - 1997/10/10

Y1 - 1997/10/10

N2 - We consider the problem of learning deterministic even linear languages from positive examples. We show that, for any nonnegative integer k, the class of LR(k) even linear languages is not learnable from positive examples while there is a subclass called LRS(k), which is a natural subclass of LR(k) in the strong sense, learnable from positive examples. Our learning algorithm identifies this subclass in the limit with almost linear time in updating conjectures. As a corollary, in terms of even linear grammars, we have a learning algorithm for k-reversible languages that is more efficient than the one proposed by Angluin.

AB - We consider the problem of learning deterministic even linear languages from positive examples. We show that, for any nonnegative integer k, the class of LR(k) even linear languages is not learnable from positive examples while there is a subclass called LRS(k), which is a natural subclass of LR(k) in the strong sense, learnable from positive examples. Our learning algorithm identifies this subclass in the limit with almost linear time in updating conjectures. As a corollary, in terms of even linear grammars, we have a learning algorithm for k-reversible languages that is more efficient than the one proposed by Angluin.

UR - http://www.scopus.com/inward/record.url?scp=0031251289&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031251289&partnerID=8YFLogxK

U2 - 10.1016/S0304-3975(97)00016-9

DO - 10.1016/S0304-3975(97)00016-9

M3 - Article

AN - SCOPUS:0031251289

VL - 185

SP - 63

EP - 79

JO - Theoretical Computer Science

JF - Theoretical Computer Science

SN - 0304-3975

IS - 1

ER -