Optimization and Systems Theory Seminar
November 9, at 11.00, room 3721, Lindstedtsvägen 25, KTH:
Giulio Bottegal, Department of Information Engineering,
University of Padova
Regularized spectrum estimation using stable spline kernels
In this talk we present a new regularized kernel based approach for
the estimation of the second order moments of stationary stochastic
processes. The correlation functions are assumed to be summable and
estimated as the solution of a Tikhonov-type variational problem. The
hypothesis space is a Reproducing kernel Hilbert space induced by the
recently introduced Stable Spline kernel. In this way, the information
on the decay to zero of the functions to be econstructed is
incorporated in the estimation process. We show that the overall
complexity of the proposed estimator scales linearly with the number
of available samples of the processes. An application to the
identification of transfer functions in the case of white noise as
input is also presented. Numerical simulations show that the proposed
method compares favorably with respect to standard nonparametric
estimation algorithms that exploit an oracle-type tuning of the
parameters.
Calendar of seminars
Last update: November 5, 2012.