| M. Davy, A. Gretton, A. Doucet, and P.W.J. Rayner, "Optimised support vector machines for nonstationary signal classification," IEEE transactions on Signal Processing, vol. 9, no. 12, Dec. 2002. |
....kernels proposed by Vapnik in [24] Ch. 11. or the time frequency kernel of 4 Data Tool d=2 d=3 d=4 d=5 AR model AR 0.640 0.624 0.624 0.633 SVM 0.648 0.634 0.632 0.639 AR outliers AR 0.942 0.922 0.919 0.911 SVM 0.885 0.826 0.832 0.827 Figure 1: Performance of AR and SVM model. Davy et al. [7]. Vapnik s kernel is based on the typical kernel trick: While the constructing the Fourier series expansion of a time series is hard, calculating the inner product in the feature space given by the Fourier expansion of order N is easy: K(x i ; x j ) sin( 2N 1 (x i x j ) sin (x i x j ) ....
M. Davy, A. Gretton, A. Doucet, and P.W.J. Rayner. Optimised support vector machines for nonstationary signal classi cation. IEEE transactions on Signal Processing, 9(12), Dec. 2002.
No context found.
M. Davy, A. Gretton, A. Doucet, and P. Rayner, "Optimised support vector machines for nonstationary signal classification," IEEE Signal Processing letters, vol. 9, no. 12, pp. 442--445, December 2002.
No context found.
M. Davy, A. Gretton, A. Doucet, and P. Rayner, "Optimised Support Vector Machines for Nonstationary Signal Classification," IEEE Signal Processing Letters, vol. 9, no. 12, pp. 442--445, Dec. 2002.
No context found.
M. Davy, A. Gretton, A. Doucet, and P.W.J. Rayner, "Optimised support vector machines for nonstationary signal classification," IEEE transactions on Signal Processing, vol. 9, no. 12, Dec. 2002.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC