Binet-Cauchy Kernels on Dynamical Systems and its Application to the Analysis of Dynamic Scenes ∗ S.V.N. Vishwanathan Statistical Machine Learning Program National ICT Australia Canberra, 0200 ACT, Australia Tel: +61 (2) 6125 8657 E-mail:
[email protected] Alexander J. Smola Statistical Machine Learning Program National ICT Australia Canberra, 0200 ACT, Australia Tel: +61 (2) 6125 8652 E-mail:
[email protected] Ren´eVidal Center for Imaging Science, Department of Biomedical Engineering, Johns Hopkins University 308B Clark Hall, 3400 N. Charles St., Baltimore MD 21218 Tel: +1 (410) 516 7306 E-mail:
[email protected] October 5, 2010 Abstract. We propose a family of kernels based on the Binet-Cauchy theorem, and its extension to Fredholm operators. Our derivation provides a unifying framework for all kernels on dynamical systems currently used in machine learning, including kernels derived from the behavioral framework, diffusion processes, marginalized kernels, kernels on graphs, and the kernels on sets arising from the subspace angle approach. In the case of linear time-invariant systems, we derive explicit formulae for computing the proposed Binet-Cauchy kernels by solving Sylvester equations, and relate the proposed kernels to existing kernels based on cepstrum coefficients and subspace angles. We show efficient methods for computing our kernels which make them viable for the practitioner. Besides their theoretical appeal, these kernels can be used efficiently in the comparison of video sequences of dynamic scenes that can be modeled as the output of a linear time-invariant dynamical system. One advantage of our kernels is that they take the initial conditions of the dynamical systems into account.