Cross-Validation Optimization for Structured Hessian Kernel Methods

Seeger, M

Local TagsRelease HistoryDetailsSummary

Cross-Validation Optimization for Structured Hessian Kernel Methods

Seeger, M.(2006). Cross-Validation Optimization for Structured Hessian Kernel Methods. Tübingen, Germany: Max Planck Institute for Biological Cybernetics.

Item is Released

show all hide all

Basic

show hide

Item Permalink: https://hdl.handle.net/11858/00-001M-0000-0013-D2F3-5 Version Permalink: https://hdl.handle.net/21.11116/0000-0002-87BC-2

Genre: Report

Files

show Files

hide Files

:

kerlogregr_3863[0].pdf (Publisher version), 367KB

View Save

File Permalink:
https://hdl.handle.net/21.11116/0000-0002-87BD-1

Name:
kerlogregr_3863[0].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:
Seeger, M^{1, 2}, Author

Affiliations:
1Department Empirical Inference, Max Planck Institute for Biological Cybernetics, Max Planck Society, ou_1497795
2Max Planck Institute for Biological Cybernetics, Max Planck Society, Spemannstrasse 38, 72076 Tübingen, DE, ou_1497794

Content

show

hide

Free keywords: -

Abstract: We address the problem of learning hyperparameters in kernel methods for
which the Hessian of the objective is structured. We propose an approximation
to the cross-validation log likelihood whose gradient can be computed
analytically, solving the hyperparameter learning problem efficiently
through nonlinear optimization. Crucially, our learning method is based
entirely on matrix-vector multiplication primitives with the kernel
matrices and their derivatives, allowing straightforward specialization to
new kernels or to large datasets. When applied to the problem of multi-way
classification, our method scales linearly in the number of classes and
gives rise to state-of-the-art results on a remote imaging task.

Details

show

hide

Language(s):

Dates: Date issued: 2006-02

Publication Status: Issued

Pages: 36

Publishing info: Tübingen, Germany : Max Planck Institute for Biological Cybernetics

Table of Contents: -

Rev. Type: -

Identifiers: BibTex Citekey: 3863

Degree: -

Event

show

Legal Case

show

Project information

show

Source 1

show

hide

Title: Technical Report of the Max Planck Institute for Biological Cybernetics

Source Genre: Series

Creator(s):

Affiliations:

Publ. Info: -

Pages: - Volume / Issue: - Sequence Number: - Start / End Page: - Identifier: -