Learning languages by collecting cases and tuning parameters

Yasubumi Sakakibara, Klaus P. Jantke, Steffen Lange

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

4 Citations (Scopus)

Abstract

We investigate the problem of case-based learning of formal languages. Case-based reasoning and learning is a currently booming area of artificial intelligence. The formal framework for case-based learning of languages has recently been developed by [JL93] in an inductive inference manner. In this paper, we first show that any indexed class of recursive languages in which finiteness is decidable is case-based representable, but many classes of languages including the class of all regular languages are not case-based learnable with a fixed universal similarity measure, even if both positive and negative examples are presented. Next we consider a framework of case-based learning where the learning algorithm is allowed to learn similarity measures, too. To avoid trivial encoding tricks, we carefully examine to what extent the similarity measure is going to be learned. Then by allowing only to learn a few parameters in the similarity measures, we show that any indexed class of recursive languages whose finiteness problem is decidable is case-based learnable. This implies that all context-free languages are case-based learnable by collecting cases and learning parameters of the similarity measures.

Original languageEnglish
Title of host publicationAlgorithmic Learning Theory - 4th International Workshop on Analogical and Inductive Inference, AII 1994 and 5th International Workshop on Algorithmic Learning Theory, ALT 1994, Proceedings
PublisherSpringer Verlag
Pages532-546
Number of pages15
Volume872 LNAI
ISBN (Print)9783540585206
Publication statusPublished - 1994
Externally publishedYes
Event4th International Workshop on Analogical and Inductive Inference, AII 1994 and 5th International Workshop on Algorithmic Learning Theory, ALT 1994 - Reinhardsbrunn Castle, Germany
Duration: 1994 Oct 101994 Oct 15

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume872 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other4th International Workshop on Analogical and Inductive Inference, AII 1994 and 5th International Workshop on Algorithmic Learning Theory, ALT 1994
CountryGermany
CityReinhardsbrunn Castle
Period94/10/1094/10/15

Fingerprint

Formal languages
Parameter Tuning
Similarity Measure
Tuning
Context free languages
Case based reasoning
Learning algorithms
Artificial intelligence
Finiteness
Inductive Inference
Parameter Learning
Context-free Languages
Formal Languages
Case-based Reasoning
Regular Languages
Learning Algorithm
Artificial Intelligence
Trivial
Encoding
Language Acquisition

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Sakakibara, Y., Jantke, K. P., & Lange, S. (1994). Learning languages by collecting cases and tuning parameters. In Algorithmic Learning Theory - 4th International Workshop on Analogical and Inductive Inference, AII 1994 and 5th International Workshop on Algorithmic Learning Theory, ALT 1994, Proceedings (Vol. 872 LNAI, pp. 532-546). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 872 LNAI). Springer Verlag.

Learning languages by collecting cases and tuning parameters. / Sakakibara, Yasubumi; Jantke, Klaus P.; Lange, Steffen.

Algorithmic Learning Theory - 4th International Workshop on Analogical and Inductive Inference, AII 1994 and 5th International Workshop on Algorithmic Learning Theory, ALT 1994, Proceedings. Vol. 872 LNAI Springer Verlag, 1994. p. 532-546 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 872 LNAI).

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

Sakakibara, Y, Jantke, KP & Lange, S 1994, Learning languages by collecting cases and tuning parameters. in Algorithmic Learning Theory - 4th International Workshop on Analogical and Inductive Inference, AII 1994 and 5th International Workshop on Algorithmic Learning Theory, ALT 1994, Proceedings. vol. 872 LNAI, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 872 LNAI, Springer Verlag, pp. 532-546, 4th International Workshop on Analogical and Inductive Inference, AII 1994 and 5th International Workshop on Algorithmic Learning Theory, ALT 1994, Reinhardsbrunn Castle, Germany, 94/10/10.
Sakakibara Y, Jantke KP, Lange S. Learning languages by collecting cases and tuning parameters. In Algorithmic Learning Theory - 4th International Workshop on Analogical and Inductive Inference, AII 1994 and 5th International Workshop on Algorithmic Learning Theory, ALT 1994, Proceedings. Vol. 872 LNAI. Springer Verlag. 1994. p. 532-546. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
Sakakibara, Yasubumi ; Jantke, Klaus P. ; Lange, Steffen. / Learning languages by collecting cases and tuning parameters. Algorithmic Learning Theory - 4th International Workshop on Analogical and Inductive Inference, AII 1994 and 5th International Workshop on Algorithmic Learning Theory, ALT 1994, Proceedings. Vol. 872 LNAI Springer Verlag, 1994. pp. 532-546 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{8b1f1babd7ef4396b4e14c09c9227d1f,
title = "Learning languages by collecting cases and tuning parameters",
abstract = "We investigate the problem of case-based learning of formal languages. Case-based reasoning and learning is a currently booming area of artificial intelligence. The formal framework for case-based learning of languages has recently been developed by [JL93] in an inductive inference manner. In this paper, we first show that any indexed class of recursive languages in which finiteness is decidable is case-based representable, but many classes of languages including the class of all regular languages are not case-based learnable with a fixed universal similarity measure, even if both positive and negative examples are presented. Next we consider a framework of case-based learning where the learning algorithm is allowed to learn similarity measures, too. To avoid trivial encoding tricks, we carefully examine to what extent the similarity measure is going to be learned. Then by allowing only to learn a few parameters in the similarity measures, we show that any indexed class of recursive languages whose finiteness problem is decidable is case-based learnable. This implies that all context-free languages are case-based learnable by collecting cases and learning parameters of the similarity measures.",
author = "Yasubumi Sakakibara and Jantke, {Klaus P.} and Steffen Lange",
year = "1994",
language = "English",
isbn = "9783540585206",
volume = "872 LNAI",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "532--546",
booktitle = "Algorithmic Learning Theory - 4th International Workshop on Analogical and Inductive Inference, AII 1994 and 5th International Workshop on Algorithmic Learning Theory, ALT 1994, Proceedings",
address = "Germany",

}

TY - GEN

T1 - Learning languages by collecting cases and tuning parameters

AU - Sakakibara, Yasubumi

AU - Jantke, Klaus P.

AU - Lange, Steffen

PY - 1994

Y1 - 1994

N2 - We investigate the problem of case-based learning of formal languages. Case-based reasoning and learning is a currently booming area of artificial intelligence. The formal framework for case-based learning of languages has recently been developed by [JL93] in an inductive inference manner. In this paper, we first show that any indexed class of recursive languages in which finiteness is decidable is case-based representable, but many classes of languages including the class of all regular languages are not case-based learnable with a fixed universal similarity measure, even if both positive and negative examples are presented. Next we consider a framework of case-based learning where the learning algorithm is allowed to learn similarity measures, too. To avoid trivial encoding tricks, we carefully examine to what extent the similarity measure is going to be learned. Then by allowing only to learn a few parameters in the similarity measures, we show that any indexed class of recursive languages whose finiteness problem is decidable is case-based learnable. This implies that all context-free languages are case-based learnable by collecting cases and learning parameters of the similarity measures.

AB - We investigate the problem of case-based learning of formal languages. Case-based reasoning and learning is a currently booming area of artificial intelligence. The formal framework for case-based learning of languages has recently been developed by [JL93] in an inductive inference manner. In this paper, we first show that any indexed class of recursive languages in which finiteness is decidable is case-based representable, but many classes of languages including the class of all regular languages are not case-based learnable with a fixed universal similarity measure, even if both positive and negative examples are presented. Next we consider a framework of case-based learning where the learning algorithm is allowed to learn similarity measures, too. To avoid trivial encoding tricks, we carefully examine to what extent the similarity measure is going to be learned. Then by allowing only to learn a few parameters in the similarity measures, we show that any indexed class of recursive languages whose finiteness problem is decidable is case-based learnable. This implies that all context-free languages are case-based learnable by collecting cases and learning parameters of the similarity measures.

UR - http://www.scopus.com/inward/record.url?scp=84981185842&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84981185842&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84981185842

SN - 9783540585206

VL - 872 LNAI

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

SP - 532

EP - 546

BT - Algorithmic Learning Theory - 4th International Workshop on Analogical and Inductive Inference, AII 1994 and 5th International Workshop on Algorithmic Learning Theory, ALT 1994, Proceedings

PB - Springer Verlag

ER -