Probabilistic generalization of simple grammars and its application to reinforcement learning

Takeshi Shibata, Ryo Yoshinaka, Takashi Chikayama

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

1 Citation (Scopus)

Abstract

Recently, some non-regular subclasses of context-free grammars have been found to be efficiently learnable from positive data. In order to use these efficient algorithms to infer probabilistic languages, one must take into account not only equivalences between languages but also probabilistic generalities of grammars. The probabilistic generality of a grammar G is the class of the probabilistic languages generated by probabilistic grammars constructed on G. We introduce a subclass of simple grammars (SGs), referred as to unifiable simple grammars (USGs), which is a superclass of an efficiently learnable class, right-unique simple grammars (RSGs). We show that the class of RSGs is unifiable within the class of USGs, whereas SGs and RSGs are not unifiable within the class of SGs and RSGs, respectively. We also introduce simple context-free decision processes, which are a natural extension of finite Markov decision processes and intuitively may be thought of a Markov decision process with stacks. We propose a reinforcement learning method on simple context-free decision processes, as an application of the learning and unification algorithm for RSGs from positive data.

Original languageEnglish
Title of host publicationAlgorithmic Learning Theory - 17th International Conference, ALT 2006, Proceedings
PublisherSpringer-Verlag
Pages348-362
Number of pages15
ISBN (Print)3540466495, 9783540466499
Publication statusPublished - 2006 Jan 1
Externally publishedYes
Event17th International Conference on Algorithmic Learning Theory, ALT 2006 - Barcelona, Spain
Duration: 2006 Oct 72006 Oct 10

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume4264 LNAI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other17th International Conference on Algorithmic Learning Theory, ALT 2006
CountrySpain
CityBarcelona
Period06/10/706/10/10

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'Probabilistic generalization of simple grammars and its application to reinforcement learning'. Together they form a unique fingerprint.

  • Cite this

    Shibata, T., Yoshinaka, R., & Chikayama, T. (2006). Probabilistic generalization of simple grammars and its application to reinforcement learning. In Algorithmic Learning Theory - 17th International Conference, ALT 2006, Proceedings (pp. 348-362). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 4264 LNAI). Springer-Verlag.