Extracting higher-level relationships in connectionist models

G. F. Marcus

Research output: Contribution to journalArticle

Abstract

Connectionist networks excel at extracting statistical regularities but have trouble extracting higher-order relationships. Clark and Thornton suggest that a solution to this problem might come from Elman (1993), but I argue that the success of Elman's single recurrent network is illusory, and we show that it cannot in fact represent abstract relationships that can be generalized to novel instances, undermining Clark and Thornton's key arguments.

Original languageEnglish (US)
Pages (from-to)77
Number of pages1
JournalBehavioral and Brain Sciences
Volume20
Issue number1
StatePublished - 1997

Fingerprint

Neural Networks (Computer)

ASJC Scopus subject areas

  • Ecology, Evolution, Behavior and Systematics
  • Neuroscience(all)
  • Psychology(all)
  • Neuropsychology and Physiological Psychology

Cite this

Extracting higher-level relationships in connectionist models. / Marcus, G. F.

In: Behavioral and Brain Sciences, Vol. 20, No. 1, 1997, p. 77.

Research output: Contribution to journalArticle

@article{784659486843453abf0d774cce1ce493,
title = "Extracting higher-level relationships in connectionist models",
abstract = "Connectionist networks excel at extracting statistical regularities but have trouble extracting higher-order relationships. Clark and Thornton suggest that a solution to this problem might come from Elman (1993), but I argue that the success of Elman's single recurrent network is illusory, and we show that it cannot in fact represent abstract relationships that can be generalized to novel instances, undermining Clark and Thornton's key arguments.",
author = "Marcus, {G. F.}",
year = "1997",
language = "English (US)",
volume = "20",
pages = "77",
journal = "Behavioral and Brain Sciences",
issn = "0140-525X",
publisher = "Cambridge University Press",
number = "1",

}

TY - JOUR

T1 - Extracting higher-level relationships in connectionist models

AU - Marcus, G. F.

PY - 1997

Y1 - 1997

N2 - Connectionist networks excel at extracting statistical regularities but have trouble extracting higher-order relationships. Clark and Thornton suggest that a solution to this problem might come from Elman (1993), but I argue that the success of Elman's single recurrent network is illusory, and we show that it cannot in fact represent abstract relationships that can be generalized to novel instances, undermining Clark and Thornton's key arguments.

AB - Connectionist networks excel at extracting statistical regularities but have trouble extracting higher-order relationships. Clark and Thornton suggest that a solution to this problem might come from Elman (1993), but I argue that the success of Elman's single recurrent network is illusory, and we show that it cannot in fact represent abstract relationships that can be generalized to novel instances, undermining Clark and Thornton's key arguments.

UR - http://www.scopus.com/inward/record.url?scp=0030854574&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0030854574&partnerID=8YFLogxK

M3 - Article

VL - 20

SP - 77

JO - Behavioral and Brain Sciences

JF - Behavioral and Brain Sciences

SN - 0140-525X

IS - 1

ER -