The Cauchy-Schwarz divergence for poisson point processes

Information theoretic divergences are fundamental tools used to measure the difference between the information conveyed by two random processes. In this paper, we show that the Cauchy-Schwarz divergence between two Poisson point processes is the half the squared L2-distance between their respective...

Full description

Bibliographic Details
Main Authors: Hoang, Hung Gia, Vo, Ba-Ngu, Vo, Ba Tuong, Mahler, R.
Format: Conference Paper
Published: IEEE Computer Society 2014
Online Access:http://hdl.handle.net/20.500.11937/14417
Description
Summary:Information theoretic divergences are fundamental tools used to measure the difference between the information conveyed by two random processes. In this paper, we show that the Cauchy-Schwarz divergence between two Poisson point processes is the half the squared L2-distance between their respective intensity functions. Moreover, this can be evaluated in closed form when the intensities are Gaussian mixtures. © 2014 IEEE.