Modeling unsupervised perceptual category learning

Brenden Lake, Gautam K. Vallabha, James L. McClelland

Research output: Contribution to journalArticle

Abstract

During the learning of speech sounds and other perceptual categories, category labels are not provided, the number of categories is unknown, and the stimuli are encountered sequentially. These constraints provide a challenge for models, but they have been recently addressed in the online mixture estimation model of unsupervised vowel category learning (see Vallabha in the reference section). The model treats categories as Gaussian distributions, proposing both the number and the parameters of the categories. While the model has been shown to successfully learn vowel categories, it has not been evaluated as a model of the learning process. We account for several results: acquired distinctiveness between categories and acquired similarity within categories, a faster increase in discrimination for more acoustically dissimilar vowels, and gradual unsupervised learning of category structure in simple visual stimuli.

Original languageEnglish (US)
Article number4895218
Pages (from-to)35-43
Number of pages9
JournalIEEE Transactions on Autonomous Mental Development
Volume1
Issue number1
DOIs
StatePublished - May 2009

Fingerprint

Unsupervised learning
Gaussian distribution
Labels
Acoustic waves

Keywords

  • Human learning
  • Mixture of Gaussians
  • Online learning
  • Unsupervised learning

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence

Cite this

Modeling unsupervised perceptual category learning. / Lake, Brenden; Vallabha, Gautam K.; McClelland, James L.

In: IEEE Transactions on Autonomous Mental Development, Vol. 1, No. 1, 4895218, 05.2009, p. 35-43.

Research output: Contribution to journalArticle

Lake, Brenden ; Vallabha, Gautam K. ; McClelland, James L. / Modeling unsupervised perceptual category learning. In: IEEE Transactions on Autonomous Mental Development. 2009 ; Vol. 1, No. 1. pp. 35-43.
@article{6eef3da3111044ec82f328ac6d17dffe,
title = "Modeling unsupervised perceptual category learning",
abstract = "During the learning of speech sounds and other perceptual categories, category labels are not provided, the number of categories is unknown, and the stimuli are encountered sequentially. These constraints provide a challenge for models, but they have been recently addressed in the online mixture estimation model of unsupervised vowel category learning (see Vallabha in the reference section). The model treats categories as Gaussian distributions, proposing both the number and the parameters of the categories. While the model has been shown to successfully learn vowel categories, it has not been evaluated as a model of the learning process. We account for several results: acquired distinctiveness between categories and acquired similarity within categories, a faster increase in discrimination for more acoustically dissimilar vowels, and gradual unsupervised learning of category structure in simple visual stimuli.",
keywords = "Human learning, Mixture of Gaussians, Online learning, Unsupervised learning",
author = "Brenden Lake and Vallabha, {Gautam K.} and McClelland, {James L.}",
year = "2009",
month = "5",
doi = "10.1109/TAMD.2009.2021703",
language = "English (US)",
volume = "1",
pages = "35--43",
journal = "IEEE Transactions on Cognitive and Developmental Systems",
issn = "2379-8920",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "1",

}

TY - JOUR

T1 - Modeling unsupervised perceptual category learning

AU - Lake, Brenden

AU - Vallabha, Gautam K.

AU - McClelland, James L.

PY - 2009/5

Y1 - 2009/5

N2 - During the learning of speech sounds and other perceptual categories, category labels are not provided, the number of categories is unknown, and the stimuli are encountered sequentially. These constraints provide a challenge for models, but they have been recently addressed in the online mixture estimation model of unsupervised vowel category learning (see Vallabha in the reference section). The model treats categories as Gaussian distributions, proposing both the number and the parameters of the categories. While the model has been shown to successfully learn vowel categories, it has not been evaluated as a model of the learning process. We account for several results: acquired distinctiveness between categories and acquired similarity within categories, a faster increase in discrimination for more acoustically dissimilar vowels, and gradual unsupervised learning of category structure in simple visual stimuli.

AB - During the learning of speech sounds and other perceptual categories, category labels are not provided, the number of categories is unknown, and the stimuli are encountered sequentially. These constraints provide a challenge for models, but they have been recently addressed in the online mixture estimation model of unsupervised vowel category learning (see Vallabha in the reference section). The model treats categories as Gaussian distributions, proposing both the number and the parameters of the categories. While the model has been shown to successfully learn vowel categories, it has not been evaluated as a model of the learning process. We account for several results: acquired distinctiveness between categories and acquired similarity within categories, a faster increase in discrimination for more acoustically dissimilar vowels, and gradual unsupervised learning of category structure in simple visual stimuli.

KW - Human learning

KW - Mixture of Gaussians

KW - Online learning

KW - Unsupervised learning

UR - http://www.scopus.com/inward/record.url?scp=77955673985&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77955673985&partnerID=8YFLogxK

U2 - 10.1109/TAMD.2009.2021703

DO - 10.1109/TAMD.2009.2021703

M3 - Article

AN - SCOPUS:77955673985

VL - 1

SP - 35

EP - 43

JO - IEEE Transactions on Cognitive and Developmental Systems

JF - IEEE Transactions on Cognitive and Developmental Systems

SN - 2379-8920

IS - 1

M1 - 4895218

ER -