Notes on Differential Entropy Calculation Using Particles
2008 (English)Report (Other academic)
This report outlines a method to calculate the differential entropy of a probability density represented by a number of particles and weights. When only the particles and the weights are given, entropy calculation is cumbersome and requires continuous approximation of the density by using some kernel functions. However, in a particle ltering framework, Bayes rule provides a direct sample based approximation to the problem.
Place, publisher, year, edition, pages
Linköping: Linköping University Electronic Press, 2008. , 8 p.
LiTH-ISY-R, ISSN 1400-3902 ; 2857
Entropy, Particle filter
IdentifiersURN: urn:nbn:se:liu:diva-56173ISRN: LiTH-ISY-R-2857OAI: oai:DiVA.org:liu-56173DiVA: diva2:316998