Neural Network Ensembles

Lars Kai Hansen; Peter Salamon

Neural Network Ensembles

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

2785 Downloads (Pure)

Abstract

We propose several means for improving the performance an training of neural networks for classification. We use crossvalidation as a tool for optimizing network parameters and architecture. We show further that the remaining generalization error can be reduced by invoking ensembles of similar networks.

Original language	English
Journal	I E E E Transactions on Pattern Analysis and Machine Intelligence
Volume	12
Pages (from-to)	993-1001
ISSN	0162-8828
Publication status	Published - 1990

Keywords

fault tolerant computing
neural networks
crossvalidation

Access to Document

Imm4833Final published version, 832 KB

OpenUrl availability

Full text

Cite this

@article{492f6c68703a4b6d97bb8509d817d00f,

title = "Neural Network Ensembles",

abstract = "We propose several means for improving the performance an training of neural networks for classification. We use crossvalidation as a tool for optimizing network parameters and architecture. We show further that the remaining generalization error can be reduced by invoking ensembles of similar networks.",

keywords = "fault tolerant computing, neural networks, crossvalidation",

author = "Hansen, {Lars Kai} and Peter Salamon",

year = "1990",

language = "English",

volume = "12",

pages = "993--1001",

journal = "I E E E Transactions on Pattern Analysis and Machine Intelligence",

issn = "0162-8828",

publisher = "Institute of Electrical and Electronics Engineers",

}

TY - JOUR

T1 - Neural Network Ensembles

AU - Hansen, Lars Kai

AU - Salamon, Peter

PY - 1990

Y1 - 1990

N2 - We propose several means for improving the performance an training of neural networks for classification. We use crossvalidation as a tool for optimizing network parameters and architecture. We show further that the remaining generalization error can be reduced by invoking ensembles of similar networks.

AB - We propose several means for improving the performance an training of neural networks for classification. We use crossvalidation as a tool for optimizing network parameters and architecture. We show further that the remaining generalization error can be reduced by invoking ensembles of similar networks.

KW - fault tolerant computing

KW - neural networks

KW - crossvalidation

M3 - Journal article

SN - 0162-8828

VL - 12

SP - 993

EP - 1001

JO - I E E E Transactions on Pattern Analysis and Machine Intelligence

JF - I E E E Transactions on Pattern Analysis and Machine Intelligence

ER -

Neural Network Ensembles

Abstract

Keywords

Access to Document

OpenUrl availability

Fingerprint

Cite this