Properties of learning in ARTMAP

Michael Georgiopoulos; Juxin Huang; Gregory L. Heileman

doi:10.1016/0893-6080(94)90083-3

Properties of learning in ARTMAP

Michael Georgiopoulos, Juxin Huang, Gregory L. Heileman

Research output: Contribution to journal › Article › peer-review

19 Scopus citations

Abstract

In this paper we consider the ARTMAP architecture for situations requiring learning of many-to-one maps. It is shown that if ARTMAP is repeatedly presented with a list of input/output pairs, it establishes the required mapping in at most M_a-1 list presentations, where M_a corresponds to the total number of ones in each one of the input patterns. Other useful properties, associated with the learning of the mapping represented by an arbitrary list of input/output pairs, are also examined. These properties reveal some of the characteristics of learning in ARTMAP when it is used as a tool in establishing an arbitrary mapping from a binary input space to a binary output space. The results presented in this paper are valid for the fast learning case, and for small β_a values, where β_a is a parameter associated with the adaptation of bottom-up weights in one of the ART1 modules of ARTMAP.

Original language	English (US)
Pages (from-to)	495-506
Number of pages	12
Journal	Neural Networks
Volume	7
Issue number	3
DOIs	https://doi.org/10.1016/0893-6080(94)90083-3
State	Published - 1994
Externally published	Yes

Keywords

ART1
ARTMAP
Adaptive resonance theory
Learning
Neural network
Pattern recognition

ASJC Scopus subject areas

Cognitive Neuroscience
Artificial Intelligence

Access to Document

10.1016/0893-6080(94)90083-3

Cite this

@article{832f138b57a443958089f1f28e6a2b99,

title = "Properties of learning in ARTMAP",

abstract = "In this paper we consider the ARTMAP architecture for situations requiring learning of many-to-one maps. It is shown that if ARTMAP is repeatedly presented with a list of input/output pairs, it establishes the required mapping in at most Ma-1 list presentations, where Ma corresponds to the total number of ones in each one of the input patterns. Other useful properties, associated with the learning of the mapping represented by an arbitrary list of input/output pairs, are also examined. These properties reveal some of the characteristics of learning in ARTMAP when it is used as a tool in establishing an arbitrary mapping from a binary input space to a binary output space. The results presented in this paper are valid for the fast learning case, and for small βa values, where βa is a parameter associated with the adaptation of bottom-up weights in one of the ART1 modules of ARTMAP.",

keywords = "ART1, ARTMAP, Adaptive resonance theory, Learning, Neural network, Pattern recognition",

author = "Michael Georgiopoulos and Juxin Huang and Heileman, {Gregory L.}",

note = "Funding Information: Acknowledgements: This research was supported in part by a grant from the Florida High Technology and Industry Council, in part by a grant from the Division of Sponsored Research at the University of Central Florida, and in part by a grant from Boeing Computer Services under contract W-300445.",

year = "1994",

doi = "10.1016/0893-6080(94)90083-3",

language = "English (US)",

volume = "7",

pages = "495--506",

journal = "Neural Networks",

issn = "0893-6080",

publisher = "Elsevier Ltd",

number = "3",

}

TY - JOUR

T1 - Properties of learning in ARTMAP

AU - Georgiopoulos, Michael

AU - Huang, Juxin

AU - Heileman, Gregory L.

N1 - Funding Information: Acknowledgements: This research was supported in part by a grant from the Florida High Technology and Industry Council, in part by a grant from the Division of Sponsored Research at the University of Central Florida, and in part by a grant from Boeing Computer Services under contract W-300445.

PY - 1994

Y1 - 1994

N2 - In this paper we consider the ARTMAP architecture for situations requiring learning of many-to-one maps. It is shown that if ARTMAP is repeatedly presented with a list of input/output pairs, it establishes the required mapping in at most Ma-1 list presentations, where Ma corresponds to the total number of ones in each one of the input patterns. Other useful properties, associated with the learning of the mapping represented by an arbitrary list of input/output pairs, are also examined. These properties reveal some of the characteristics of learning in ARTMAP when it is used as a tool in establishing an arbitrary mapping from a binary input space to a binary output space. The results presented in this paper are valid for the fast learning case, and for small βa values, where βa is a parameter associated with the adaptation of bottom-up weights in one of the ART1 modules of ARTMAP.

AB - In this paper we consider the ARTMAP architecture for situations requiring learning of many-to-one maps. It is shown that if ARTMAP is repeatedly presented with a list of input/output pairs, it establishes the required mapping in at most Ma-1 list presentations, where Ma corresponds to the total number of ones in each one of the input patterns. Other useful properties, associated with the learning of the mapping represented by an arbitrary list of input/output pairs, are also examined. These properties reveal some of the characteristics of learning in ARTMAP when it is used as a tool in establishing an arbitrary mapping from a binary input space to a binary output space. The results presented in this paper are valid for the fast learning case, and for small βa values, where βa is a parameter associated with the adaptation of bottom-up weights in one of the ART1 modules of ARTMAP.

KW - ART1

KW - ARTMAP

KW - Adaptive resonance theory

KW - Learning

KW - Neural network

KW - Pattern recognition

UR - http://www.scopus.com/inward/record.url?scp=0028312869&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0028312869&partnerID=8YFLogxK

U2 - 10.1016/0893-6080(94)90083-3

DO - 10.1016/0893-6080(94)90083-3

M3 - Article

AN - SCOPUS:0028312869

SN - 0893-6080

VL - 7

SP - 495

EP - 506

JO - Neural Networks

JF - Neural Networks

IS - 3

ER -

Properties of learning in ARTMAP

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this