Efficient and Accurate Lp-Norm Multiple Kernel Learning

Kloft, M; Brefeld, U; Sonnenburg, S; Laskov, P; Müller, K-R; Zien, A

Local TagsRelease HistoryDetailsSummary

Efficient and Accurate Lp-Norm Multiple Kernel Learning

Kloft, M., Brefeld, U., Sonnenburg, S., Laskov, P., Müller, K.-R., & Zien, A. (2010). Efficient and Accurate Lp-Norm Multiple Kernel Learning. In Y. Bengio, D. Schuurmans, J. Lafferty, C. Williams, & A. Culotta (Eds.), Advances in Neural Information Processing Systems 22 (pp. 997-1005). Red Hook, NY, USA: Curran.

Item is Released

show all

Basic

hide

Item Permalink: https://hdl.handle.net/21.11116/0000-0010-841C-A Version Permalink: https://hdl.handle.net/21.11116/0000-0010-841D-9

Genre: Conference Paper

Files

show Files

Locators

hide

Locator:
https://papers.nips.cc/paper_files/paper/2009/hash/d516b13671a4179d9b7b458a6ebdeb92-Abstract.html (Publisher version) Open Access status unknown

Description:
-

OA-Status:
Not specified

Creators

hide

Creators:
Kloft, M, Author
Brefeld, U, Author
Sonnenburg, S¹, Author
Laskov, P, Author
Müller, K-R, Author
Zien, A, Author

Affiliations:
1Rätsch Group, Friedrich Miescher Laboratory, Max Planck Society, ou_3378052

Content

hide

Free keywords: -

Abstract: Learning linear combinations of multiple kernels is an appealing strategy when the right choice of features is unknown. Previous approaches to multiple kernel learning (MKL) promote sparse kernel combinations and hence support interpretability. Unfortunately, L1-norm MKL is hardly observed to outperform trivial baselines in practical applications. To allow for robust kernel mixtures, we generalize MKL to arbitrary Lp-norms. We devise new insights on the connection between several existing MKL formulations and develop two efficient interleaved optimization strategies for arbitrary p>1. Empirically, we demonstrate that the interleaved optimization strategies are much faster compared to the traditionally used wrapper approaches. Finally, we apply Lp-norm MKL to real-world problems from computational biology, showing that non-sparse MKL achieves accuracies that go beyond the state-of-the-art.

Details

hide

Language(s):

Dates: Date issued: 2010-04

Publication Status: Issued

Pages: -

Publishing info: -

Table of Contents: -

Rev. Type: -

Identifiers: -

Degree: -

Event

hide

Title: 23rd Annual Conference on Neural Information Processing Systems (NIPS 2009)

Place of Event: Vancouver, BC, Canada

Start-/End Date: 2009-12-07 - 2009-12-10

Legal Case

show

Project information

show

Source 1

hide

Title: Advances in Neural Information Processing Systems 22

Source Genre: Proceedings

Creator(s):
Bengio, Y, Editor
Schuurmans, D, Editor
Lafferty, J, Editor
Williams, C, Editor
Culotta, A, Editor

Affiliations:
-

Publ. Info: Red Hook, NY, USA : Curran

Pages: - Volume / Issue: - Sequence Number: - Start / End Page: 997 - 1005 Identifier: ISBN: 978-1-615-67911-9