Constraints on generalization by adaptive networks

M. Pavel; Holly Jimison; Rebecca T. Moore

doi:10.1016/0893-6080(88)90083-4

Constraints on generalization by adaptive networks

M. Pavel, Holly Jimison, Rebecca T. Moore

Biomedical Engineering

Research output: Contribution to journal › Conference article › peer-review

Abstract

There are three classes of constraints that can be used to optimize generalization: Constraints on the network architecture, constraints on the learning algorithms, and constraints imposed by the representation of inputs and outputs. In this paper we focus on the constraints imposed by the architecture. These constraints are related to the topology of the networks, limitations on the number of units, and connectivity. In this project we used a computational approach (rather than simulation) to examine the effects of architectural constraints on generalization. Our analysis is restricted to two-layer non-recurrent networks with linear threshold units. Using this approach we characterized the effects of a variety of constraints, including minimizing the number of hidden units.

Original language	English (US)
Number of pages	1
Journal	Neural Networks
Volume	1
Issue number	1 SUPPL
DOIs	https://doi.org/10.1016/0893-6080(88)90083-4
State	Published - Jan 1 1988
Event	International Neural Network Society 1988 First Annual Meeting - Boston, MA, USA Duration: Sep 6 1988 → Sep 10 1988

ASJC Scopus subject areas

Cognitive Neuroscience
Artificial Intelligence

Access to Document

10.1016/0893-6080(88)90083-4

Cite this

@article{31e3b5ac0482489fbc49fe7aef7f1018,

title = "Constraints on generalization by adaptive networks",

abstract = "There are three classes of constraints that can be used to optimize generalization: Constraints on the network architecture, constraints on the learning algorithms, and constraints imposed by the representation of inputs and outputs. In this paper we focus on the constraints imposed by the architecture. These constraints are related to the topology of the networks, limitations on the number of units, and connectivity. In this project we used a computational approach (rather than simulation) to examine the effects of architectural constraints on generalization. Our analysis is restricted to two-layer non-recurrent networks with linear threshold units. Using this approach we characterized the effects of a variety of constraints, including minimizing the number of hidden units.",

author = "M. Pavel and Holly Jimison and Moore, {Rebecca T.}",

year = "1988",

month = jan,

day = "1",

doi = "10.1016/0893-6080(88)90083-4",

language = "English (US)",

volume = "1",

journal = "Neural Networks",

issn = "0893-6080",

publisher = "Elsevier Limited",

number = "1 SUPPL",

note = "International Neural Network Society 1988 First Annual Meeting ; Conference date: 06-09-1988 Through 10-09-1988",

}

TY - JOUR

T1 - Constraints on generalization by adaptive networks

AU - Pavel, M.

AU - Jimison, Holly

AU - Moore, Rebecca T.

PY - 1988/1/1

Y1 - 1988/1/1

N2 - There are three classes of constraints that can be used to optimize generalization: Constraints on the network architecture, constraints on the learning algorithms, and constraints imposed by the representation of inputs and outputs. In this paper we focus on the constraints imposed by the architecture. These constraints are related to the topology of the networks, limitations on the number of units, and connectivity. In this project we used a computational approach (rather than simulation) to examine the effects of architectural constraints on generalization. Our analysis is restricted to two-layer non-recurrent networks with linear threshold units. Using this approach we characterized the effects of a variety of constraints, including minimizing the number of hidden units.

AB - There are three classes of constraints that can be used to optimize generalization: Constraints on the network architecture, constraints on the learning algorithms, and constraints imposed by the representation of inputs and outputs. In this paper we focus on the constraints imposed by the architecture. These constraints are related to the topology of the networks, limitations on the number of units, and connectivity. In this project we used a computational approach (rather than simulation) to examine the effects of architectural constraints on generalization. Our analysis is restricted to two-layer non-recurrent networks with linear threshold units. Using this approach we characterized the effects of a variety of constraints, including minimizing the number of hidden units.

UR - http://www.scopus.com/inward/record.url?scp=0024172969&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0024172969&partnerID=8YFLogxK

U2 - 10.1016/0893-6080(88)90083-4

DO - 10.1016/0893-6080(88)90083-4

M3 - Conference article

AN - SCOPUS:0024172969

SN - 0893-6080

VL - 1

JO - Neural Networks

JF - Neural Networks

IS - 1 SUPPL

T2 - International Neural Network Society 1988 First Annual Meeting

Y2 - 6 September 1988 through 10 September 1988

ER -

Constraints on generalization by adaptive networks

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this