A Nonparametric Bayesian Model of Multi-Level Category Learning

Kevin R. Canini, Thomas L. Griffiths

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

3 Scopus citations

Abstract

Categories are often organized into hierarchical taxonomies, that is, tree structures where each node represents a labeled category, and a node's parent and children are, respectively, the category's supertype and subtypes. A natural question is whether it is possible to reconstruct category taxonomies in cases where we are not given explicit information about how categories are related to each other, but only a sample of observations of the members of each category. In this paper, we introduce a nonparametric Bayesian model of multi-level category learning, an extension of the hierarchical Dirichlet process (HDP) that we call the tree-HDP. We demonstrate the ability of the tree-HDP to reconstruct simulated datasets of artificial taxonomies, and show that it produces similar performance to human learners on a taxonomy inference task.

Original languageEnglish (US)
Title of host publicationProceedings of the 25th AAAI Conference on Artificial Intelligence, AAAI 2011
PublisherAAAI press
Pages307-312
Number of pages6
ISBN (Electronic)9781577355083
StatePublished - Aug 11 2011
Externally publishedYes
Event25th AAAI Conference on Artificial Intelligence, AAAI 2011 - San Francisco, United States
Duration: Aug 7 2011Aug 11 2011

Publication series

NameProceedings of the 25th AAAI Conference on Artificial Intelligence, AAAI 2011

Conference

Conference25th AAAI Conference on Artificial Intelligence, AAAI 2011
Country/TerritoryUnited States
CitySan Francisco
Period8/7/118/11/11

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'A Nonparametric Bayesian Model of Multi-Level Category Learning'. Together they form a unique fingerprint.

Cite this