GENERALIZED RECURSIVE SPLITTING ALGORITHMS FOR LEARNING HYBRID CONCEPTS

Bruce Lambert, David Tcheng, Stephen C.Y. Lu

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

2 Scopus citations

Abstract

This paper describes the Competitive Relation Learner (CRL), a generalized recursive splitting algorithm capable of producing a wide range of hybrid concept representations through the competitive application of multiple learning strategies, multiple decomposition strategies, and multiple decomposition evaluation strategies. Experimental results are reported that demonstrate CRL's ability to outperform several well known fixed-bias strategies.

Original languageEnglish (US)
Title of host publicationProceedings of the 6th International Workshop on Machine Learning, ICML 1989
EditorsAlberto Maria Segre
PublisherMorgan Kaufmann Publishers, Inc.
Pages496-498
Number of pages3
ISBN (Electronic)1558600361, 9781558600362
DOIs
StatePublished - 1989
Externally publishedYes
Event6th International Workshop on Machine Learning, ICML 1989 - Ithaca, United States
Duration: Jun 26 1989Jun 27 1989

Publication series

NameProceedings of the 6th International Workshop on Machine Learning, ICML 1989

Conference

Conference6th International Workshop on Machine Learning, ICML 1989
Country/TerritoryUnited States
CityIthaca
Period6/26/896/27/89

Funding

This research was supported in part by the Applied Intelligent Systems Group of Digital Equipment Corporation and by the National Science Foundation (DMC-8657116).

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Software
  • Theoretical Computer Science
  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'GENERALIZED RECURSIVE SPLITTING ALGORITHMS FOR LEARNING HYBRID CONCEPTS'. Together they form a unique fingerprint.

Cite this