Title :
A new metric for probability distributions
Author :
Endres, Dominik M. ; Schindelin, Johannes E.
Author_Institution :
Sch. of Psychol., Univ. of St Andrews, UK
fDate :
7/1/2003 12:00:00 AM
Abstract :
We introduce a metric for probability distributions, which is bounded, information-theoretically motivated, and has a natural Bayesian interpretation. The square root of the well-known χ2 distance is an asymptotic approximation to it. Moreover, it is a close relative of the capacitory discrimination and Jensen-Shannon divergence.
Keywords :
Bayes methods; information theory; probability; χ2 distance; Bayesian interpretation; Jensen-Shannon divergence; asymptotic approximation; bounded information-theoretically motivated metric; capacitory discrimination; probability distributions; square root; Adaptive estimation; Algorithm design and analysis; Bayesian methods; Convergence; Gaussian noise; Iterative algorithms; Probability distribution; Wavelet analysis; White noise; Writing;
Journal_Title :
Information Theory, IEEE Transactions on
DOI :
10.1109/TIT.2003.813506
