PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Wavelet Kernel Learning
Florian Yger and Alain Rakotomamonjy
Pattern Recognition 2010.


This paper addresses the problem of optimal feature extraction from a wavelet representation. Our work aims at building features by selecting wavelet coefficients resulting from signal or image decomposition on a adapted wavelet basis. For this purpose, we jointly learn in a kernelized large-margin context the wavelet shape as well as the appropriate scale and translation of the wavelets, hence the name “wavelet kernel learning”. This problem is posed as a multiple kernel learning problem where the number of kernels can be very large. For solving such a problem, we introduce a novel multiple kernel learning algorithm based on active constraints methods. We furthermore propose some variants of this algorithm that can produce approximate solutions more efficiently. Empirical analysis show that our active constraint MKL algorithm achieves state-of-the art efficiency. When applied to wavelet kernel learning, our experimental results show that the approaches we propose are competitive with respect to the state of the art on Brain-Computer Interface and Brodatz texture datasets.

EPrint Type:Article
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Learning/Statistics & Optimisation
Brain Computer Interfaces
ID Code:7554
Deposited By:Alain Rakotomamonjy
Deposited On:17 March 2011