Electrical Engineering
      and Computer Sciences

Electrical Engineering and Computer Sciences

COLLEGE OF ENGINEERING

UC Berkeley

Bayesian Localized Multiple Kernel Learning

Mario Christoudias, Raquel Urtasun and Trevor Darrell

EECS Department
University of California, Berkeley
Technical Report No. UCB/EECS-2009-96
July 3, 2009

http://www.eecs.berkeley.edu/Pubs/TechRpts/2009/EECS-2009-96.pdf

Multiple kernel learning approaches form a set of techniques for performing classification that can easily combine information from multiple data sources, e.g., by adding or multiplying kernels. Most methods, however, are limited by their assumption of a per-view kernel weighting. For many problems, the set of features important for discriminating between examples can vary locally. As a consequence these global techniques suffer in the presence of complex noise processes, such as heteroscedastic noise, or when the discriminative properties of each feature type varies across the input space. In this paper, we propose a localized multiple kernel learning approach with Gaussian Processes that learns a local weighting over each view and can obtain accurate classification performance and deal with insufficient views corrupted by complex noise, e.g., per-sample occlusion. We demonstrate our approach on the tasks of audio-visual gesture recognition and object category classification on the Caltech-101 benchmark.


BibTeX citation:

@techreport{Christoudias:EECS-2009-96,
    Author = {Christoudias, Mario and Urtasun, Raquel and Darrell, Trevor},
    Title = {Bayesian Localized Multiple Kernel Learning},
    Institution = {EECS Department, University of California, Berkeley},
    Year = {2009},
    Month = {Jul},
    URL = {http://www.eecs.berkeley.edu/Pubs/TechRpts/2009/EECS-2009-96.html},
    Number = {UCB/EECS-2009-96},
    Abstract = {Multiple kernel learning approaches form a set of techniques for performing
classification that can easily combine information from multiple data sources, e.g.,
by adding or multiplying kernels. Most methods, however, are limited by their
assumption of a per-view kernel weighting. For many problems, the set of features
important for discriminating between examples can vary locally. As a consequence
these global techniques suffer in the presence of complex noise processes, such as
heteroscedastic noise, or when the discriminative properties of each feature type
varies across the input space. In this paper, we propose a localized multiple kernel
learning approach with Gaussian Processes that learns a local weighting over each
view and can obtain accurate classification performance and deal with insufficient
views corrupted by complex noise, e.g., per-sample occlusion. We demonstrate
our approach on the tasks of audio-visual gesture recognition and object category
classification on the Caltech-101 benchmark.}
}

EndNote citation:

%0 Report
%A Christoudias, Mario
%A Urtasun, Raquel
%A Darrell, Trevor
%T Bayesian Localized Multiple Kernel Learning
%I EECS Department, University of California, Berkeley
%D 2009
%8 July 3
%@ UCB/EECS-2009-96
%U http://www.eecs.berkeley.edu/Pubs/TechRpts/2009/EECS-2009-96.html
%F Christoudias:EECS-2009-96