UNIGE document Chapitre d'actes
previous document  unige:6892  next document
add to browser collection
Title

Margin and radius based multiple Kernel Learning

Authors
Published in Buntine, Wray, Grobelnik, Marko, Mladeni, Dunja & Shawe-Taylor, John. Machine Learning and Knowledge Discovery in Databases. Bled (Slovenia) - September 7-11 2009 - Berlin, Heidelberg: Springer. 2009, p. 330-343
Collection Lecture Notes in Computer Science; 5781
Abstract A serious drawback of kernel methods, and Support Vector Machines (SVM) in particular, is the difficulty in choosing a suitable kernel function for a given dataset. One of the approaches proposed to address this problem is Multiple Kernel Learning (MKL) in which several kernels are combined adaptively for a given dataset. Many of the existing MKL methods use the SVM objective function and try to find a linear combination of basic kernels such that the separating margin between the classes is maximized. However, these methods ignore the fact that the theoretical error bound depends not only on the margin, but also on the radius of the smallest sphere that contains all the training instances. We present a novel MKL algorithm that optimizes the error bound taking account of both the margin and the radius. The empirical results show that the proposed method compares favorably with other state-of-the-art MKL methods.
Keywords Learning Kernel CombinationSupport Vector MachinesConvex optimization
Identifiers
ISBN: 978-3-642-04179-2
Full text
Structures
Research group Geneva Artificial Intelligence Laboratory
Citation
(ISO format)
DO, Thi Thanh Huyen et al. Margin and radius based multiple Kernel Learning. In: Buntine, Wray, Grobelnik, Marko, Mladeni, Dunja & Shawe-Taylor, John (Ed.). Machine Learning and Knowledge Discovery in Databases. Bled (Slovenia). Berlin, Heidelberg : Springer, 2009. p. 330-343. (Lecture Notes in Computer Science; 5781) https://archive-ouverte.unige.ch/unige:6892

357 hits

243 downloads

Update

Deposited on : 2010-06-18

Export document
Format :
Citation style :