Ensemble construction via designed output distortion
Ensemble construction via designed output distortion
A new technique for generating regression ensembles is introduced in the present paper. The technique is based on earlier work on promoting model diversity through injection of noise into the outputs; it differs from the earlier methods in its rigorous requirement that the mean displacements applied to any data points output value be exactly zero.
It is illustrated how even the introduction of extremely large displacements may lead to prediction accuracy superior to that achieved by bagging.
It is demonstrated how ensembles of models with very high bias may have much better prediction accuracy than single models of the same bias-defying the conventional belief that ensembling high bias models is not purposeful. Finally is outlined how the technique may be applied to classification.
computer science
286-295
Christensen, Stefan W.
eec9c1f7-07c8-4a94-a06b-49c29a315c7a
2003
Christensen, Stefan W.
eec9c1f7-07c8-4a94-a06b-49c29a315c7a
Christensen, Stefan W.
(2003)
Ensemble construction via designed output distortion.
Lecture Notes in Computer Science, 2709, .
Abstract
A new technique for generating regression ensembles is introduced in the present paper. The technique is based on earlier work on promoting model diversity through injection of noise into the outputs; it differs from the earlier methods in its rigorous requirement that the mean displacements applied to any data points output value be exactly zero.
It is illustrated how even the introduction of extremely large displacements may lead to prediction accuracy superior to that achieved by bagging.
It is demonstrated how ensembles of models with very high bias may have much better prediction accuracy than single models of the same bias-defying the conventional belief that ensembling high bias models is not purposeful. Finally is outlined how the technique may be applied to classification.
This record has no associated files available for download.
More information
Published date: 2003
Keywords:
computer science
Identifiers
Local EPrints ID: 19928
URI: http://eprints.soton.ac.uk/id/eprint/19928
ISSN: 0302-9743
PURE UUID: d16f99b2-269a-4fb3-ad59-10f2e0fcdf20
Catalogue record
Date deposited: 24 Feb 2006
Last modified: 08 Jan 2022 01:00
Export record
Contributors
Author:
Stefan W. Christensen
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics