Kernel methods and the exponential family

Authors:
Stéphane Canu;Alex Smola
Affiliations:
1-PSI-FRE CNRS 2645, INSA de Rouen, France, St Etienne du Rouvray, France;Statistical Machine Learning Program, National ICT, Australia and RSISE, Australian National University, Canberra, 0200 ACT, Australia
Venue:
Neurocomputing
Year:
2006

Citing 3
Cited 11

Heteroscedastic Gaussian process regression

ICML '05 Proceedings of the 22nd international conference on Machine learning
Estimating the Support of a High-Dimensional Distribution

Neural Computation
An improved training algorithm for nonlinear kernel discriminants

IEEE Transactions on Signal Processing - Part I

Causal reasoning by evaluating the complexity of conditional densities with kernel methods

Neurocomputing
Better multiclass classification via a margin-optimized single binary problem

Pattern Recognition Letters
A novel kernel-based maximum a posteriori classification method

Neural Networks
Rough cluster algorithm based on kernel function

RSKT'08 Proceedings of the 3rd international conference on Rough sets and knowledge technology
Entropy and margin maximization for structured output learning

ECML PKDD'10 Proceedings of the 2010 European conference on Machine learning and knowledge discovery in databases: Part III
Information, Divergence and Risk for Binary Experiments

The Journal of Machine Learning Research
Column subset selection for active learning in image classification

Neurocomputing
Rough kernel clustering algorithm with adaptive parameters

AICI'11 Proceedings of the Third international conference on Artificial intelligence and computational intelligence - Volume Part III
Support vector machines with weighted regularization

ICONIP'11 Proceedings of the 18th international conference on Neural Information Processing - Volume Part II
Inhibition in multiclass classification

Neural Computation
Social coordination assessment: distinguishing between shape and timing

MPRSS'12 Proceedings of the First international conference on Multimodal Pattern Recognition of Social Signals in Human-Computer-Interaction

Quantified Score

Hi-index	0.01

Visualization

Abstract

The success of support vector machine (SVM) has given rise to the development of a new class of theoretically elegant learning machines which use a central concept of kernels and the associated reproducing kernel Hilbert space (RKHS). Exponential families, a standard tool in statistics, can be used to unify many existing machine learning algorithms based on kernels (such as SVM) and to invent novel ones quite effortlessly. A new derivation of the novelty detection algorithm based on the one class SVM is proposed to illustrate the power of the exponential family model in an RKHS.