Exploring model selection techniques for nonlinear dimensionality reduction

Harmeling, S

Local TagsRelease HistoryDetailsSummary

Exploring model selection techniques for nonlinear dimensionality reduction

Harmeling, S.(2007). Exploring model selection techniques for nonlinear dimensionality reduction (EDI-INF-RR-0960). Edinburgh, UK: School of Informatics, University of Edinburgh.

Item is Released

show all hide all

Basic

show hide

Item Permalink: https://hdl.handle.net/11858/00-001M-0000-0013-CE87-3 Version Permalink: https://hdl.handle.net/21.11116/0000-0003-ED30-C

Genre: Report

Files

show Files

hide Files

:

EDI-INF-RR-0960.pdf (Publisher version), 4MB

View Save

File Permalink:
https://hdl.handle.net/21.11116/0000-0003-ED31-B

Name:
EDI-INF-RR-0960.pdf

Description:
-

OA-Status:

Visibility:
Public

MIME-Type / Checksum:
application/pdf / [MD5]

Technical Metadata:

View

Copyright Date:
-

Copyright Info:
-

License:
-

Locators

show

Creators

show

hide

Creators:
Harmeling, S¹, Author

Affiliations:
1External Organizations, ou_persistent22

Content

show

hide

Free keywords: -

Abstract: Nonlinear dimensionality reduction (NLDR) methods have become useful tools for practitioners who are faced with the analysis of high-dimensional data. Of course, not all NLDR methods are equally applicable to a particular dataset at hand. Thus it would be useful to come up with model selection criteria that help to choose among diﬀerent NLDR algorithms. This paper explores various approaches to this problem and evaluates them on controlled data sets. Comprehensive experiments will show that model selection scores based on stability are not useful, while scores based on Gaussian processes are helpful for the NLDR problem.

Details

show

hide

Language(s):

Dates: Published in Print: 2007-03

Publication Status: Published in print

Pages: 20

Publishing info: Edinburgh, UK : School of Informatics, University of Edinburgh

Table of Contents: -

Rev. Type: -

Identifiers: Report Nr.: EDI-INF-RR-0960
BibTex Citekey: 6346

Degree: -

Event

show

Legal Case

show

Project information

show

Source

show