de.mpg.escidoc.pubman.appbase.FacesBean
English
 
Help Guide Disclaimer Contact us Login
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT

Released

Conference Paper

Learning output kernels with block coordinate descent

MPS-Authors
http://pubman.mpdl.mpg.de/cone/persons/resource/persons83886

Dinuzzo,  F
Department Empirical Inference, Max Planck Institute for Biological Cybernetics, Max Planck Society;

http://pubman.mpdl.mpg.de/cone/persons/resource/persons84118

Ong,  CS
Department Empirical Inference, Max Planck Institute for Biological Cybernetics, Max Planck Society;

http://pubman.mpdl.mpg.de/cone/persons/resource/persons44483

Gehler,  PV
Department Empirical Inference, Max Planck Institute for Biological Cybernetics, Max Planck Society;

Locator
There are no locators available
Fulltext (public)
There are no public fulltexts available
Supplementary Material (public)
There is no public supplementary material available
Citation

Dinuzzo, F., Ong, C., Gehler, P., & Pillonetto, G. (2011). Learning output kernels with block coordinate descent. In 28th International Conference on Machine Learning (ICML 2011) (pp. 49-56). Madison, WI, USA: International Machine Learning Society.


Cite as: http://hdl.handle.net/11858/00-001M-0000-0013-BB22-D
Abstract
We propose a method to learn simultaneously a vector-valued function and a kernel between its components. The obtained kernel can be used both to improve learning performances and to reveal structures in the output space which may be important in their own right. Our method is based on the solution of a suitable regularization problem over a reproducing kernel Hilbert space (RKHS) of vector-valued functions. Although the regularized risk functional is non-convex, we show that it is invex, implying that all local minimizers are global minimizers. We derive a block-wise coordinate descent method that efficiently exploits the structure of the objective functional. Then, we empirically demonstrate that the proposed method can improve classification accuracy. Finally, we provide a visual interpretation of the learned kernel matrix for some well known datasets.