RUcore: Rutgers University Community Repository
Improving on-line learning
Descriptive
TitleInfo
(displayLabel = Citation Title);
(type = uniform)
Title
Improving on-line learning
Name
(ID = NAME001);
(type = personal)
NamePart
(type = family)
Mesterharm
NamePart
(type = given)
Chris
DisplayForm
Chris Mesterharm
Role
RoleTerm
(authority = RUETD)
author
Name
(ID = NAME002);
(type = personal)
NamePart
(type = family)
Hirsh
NamePart
(type = given)
Haym
Affiliation
Advisory Committee
DisplayForm
Haym Hirsh
Role
RoleTerm
(authority = RULIB)
chair
Name
(ID = NAME003);
(type = personal)
NamePart
(type = family)
Littman
NamePart
(type = given)
Michael
Affiliation
Advisory Committee
DisplayForm
Michael Littman
Role
RoleTerm
(authority = RULIB)
internal member
Name
(ID = NAME004);
(type = personal)
NamePart
(type = family)
Steiger
NamePart
(type = given)
William
Affiliation
Advisory Committee
DisplayForm
William Steiger
Role
RoleTerm
(authority = RULIB)
internal member
Name
(ID = NAME005);
(type = personal)
NamePart
(type = family)
Schapire
NamePart
(type = given)
Robert
Affiliation
Advisory Committee
DisplayForm
Robert Schapire
Role
RoleTerm
(authority = RULIB)
outside member
Name
(ID = NAME006);
(type = corporate)
NamePart
Rutgers University
Role
RoleTerm
(authority = RULIB)
degree grantor
Name
(ID = NAME007);
(type = corporate)
NamePart
Graduate School - New Brunswick
Role
RoleTerm
(authority = RULIB)
school
TypeOfResource
Text
Genre
(authority = marcgt)
theses
OriginInfo
DateCreated
(qualifier = exact)
2007
DateOther
(qualifier = exact);
(type = degree)
2007
Language
LanguageTerm
English
PhysicalDescription
Form
(authority = marcform)
electronic
InternetMediaType
application/pdf
InternetMediaType
text/xml
Extent
xv, 292 pages
Abstract
In this dissertation, we consider techniques to improve the performance and applicability of algorithms used for on-line learning. We organize these techniques according to the assumptions they make about the generation of instances. Our first assumption is that the
instances are generated by a fixed distribution. Many algorithms are designed to perform well when instances are generated by an adversary;
we give two techniques to modify these algorithms to improve performance when the instances are instead generated by a distribution. We validate these techniques with extensive experiments using a wide range of real world data sets. Our second assumption is
that the target concept the algorithm is attempting to learn changes over time. We give a modification of the Winnow algorithm and show it
has good bounds for tracking a shifting concept when instances are generated by an adversary. We also consider the case that the instances are generated by a shifting distribution. We apply
variations of the previous fixed distribution techniques and show, with real data derived experiments, that these techniques continue to
significantly improve performance. Last, we assume that the labels for instances may be delayed for a number of trials. We give techniques to modify an on-line algorithm so that it has good performance even when the labels are delayed. We derive upper-bounds on the performance of these modifications and show through lower-bounds that these modifications are close to optimal.
instances are generated by a fixed distribution. Many algorithms are designed to perform well when instances are generated by an adversary;
we give two techniques to modify these algorithms to improve performance when the instances are instead generated by a distribution. We validate these techniques with extensive experiments using a wide range of real world data sets. Our second assumption is
that the target concept the algorithm is attempting to learn changes over time. We give a modification of the Winnow algorithm and show it
has good bounds for tracking a shifting concept when instances are generated by an adversary. We also consider the case that the instances are generated by a shifting distribution. We apply
variations of the previous fixed distribution techniques and show, with real data derived experiments, that these techniques continue to
significantly improve performance. Last, we assume that the labels for instances may be delayed for a number of trials. We give techniques to modify an on-line algorithm so that it has good performance even when the labels are delayed. We derive upper-bounds on the performance of these modifications and show through lower-bounds that these modifications are close to optimal.
Note
(type = degree)
Ph.D.
Note
(type = bibliography)
Includes bibliographical references (p. 284-290).
Subject
(ID = SUBJ1);
(authority = RUETD)
Topic
Computer Science
Subject
(ID = SUBJ2);
(authority = ETD-LCSH)
Topic
Machine learning
Subject
(ID = SUBJ3);
(authority = ETD-LCSH)
Topic
Computational learning theory
RelatedItem
(type = host)
TitleInfo
Title
Graduate School - New Brunswick Electronic Theses and Dissertations
Identifier
(type = local)
rucore19991600001
Identifier
(type = hdl)
http://hdl.rutgers.edu/1782.2/rucore10001600001.ETD.16748
Identifier
ETD_428
Location
PhysicalLocation
(authority = marcorg);
(displayLabel = Rutgers, The State University of New Jersey)
NjNbRU
Identifier
(type = doi)
doi:10.7282/T34F1R43
Genre
(authority = ExL-Esploro)
ETD doctoral
Back to the top
Rights
RightsDeclaration
(AUTHORITY = GS);
(ID = rulibRdec0006)
The author owns the copyright to this work.
Copyright
Status
Copyright protected
Availability
Status
Open
AssociatedEntity
(AUTHORITY = rulib);
(ID = 1)
Name
Jon Mesterharm
Role
Copyright holder
Affiliation
Rutgers University. Graduate School - New Brunswick
RightsEvent
(AUTHORITY = rulib);
(ID = 1)
Type
Permission or license
Detail
Non-exclusive ETD license
AssociatedObject
(AUTHORITY = rulib);
(ID = 1)
Type
License
Name
Author Agreement License
Detail
I hereby grant to the Rutgers University Libraries and to my school the non-exclusive right to archive, reproduce and distribute my thesis or dissertation, in whole or in part, and/or my abstract, in whole or in part, in and from an electronic format, subject to the release date subsequently stipulated in this submittal form and approved by my school. I represent and stipulate that the thesis or dissertation and its abstract are my original work, that they do not infringe or violate any rights of others, and that I make these grants as the sole owner of the rights to my thesis or dissertation and its abstract. I represent that I have obtained written permissions, when necessary, from the owner(s) of each third party copyrighted matter to be included in my thesis or dissertation and will supply copies of such upon request by my school. I acknowledge that RU ETD and my school will not distribute my thesis or dissertation or its abstract if, in their reasonable judgment, they believe all such rights have not been secured. I acknowledge that I retain ownership rights to the copyright of my work. I also retain the right to use all or part of this thesis or dissertation in future works, such as articles or books.
Back to the top
Technical
Format
(TYPE = mime);
(VERSION = )
application/x-tar
FileSize
(UNIT = bytes)
2194432
Checksum
(METHOD = SHA1)
490b89e910b297880e84a9d443813d86af849dfc
ContentModel
ETD
CompressionScheme
other
OperatingSystem
(VERSION = 5.1)
windows xp
Format
(TYPE = mime);
(VERSION = NULL)
application/x-tar
Back to the top
Statistical Profile