Title :
Maximum entropy versus minimum risk and applications to some classical discrete distributions
Author :
Topsøe, Flemming
Author_Institution :
Dept. of Math., Copenhagen Univ., Denmark
fDate :
8/1/2002 12:00:00 AM
Abstract :
The game which can be taken to lie behind the maximum-entropy principle is studied. Refining previous techniques, new theoretical results are obtained. These results are illustrated by concrete examples pertaining to well-known classical models.
Keywords :
binary codes; game theory; maximum entropy methods; statistical analysis; Nash equilibrium; Poisson distribution; binary prefix-free code; binomial distribution; classical discrete distributions; code length game; empirical distribution; geometric distribution; maximum entropy; minimum risk; optimality criteria; Algorithm design and analysis; Artificial neural networks; Classification algorithms; Classification tree analysis; Data analysis; Entropy; Iterative algorithms; Pattern recognition; Regression tree analysis; Source coding;
Journal_Title :
Information Theory, IEEE Transactions on
DOI :
10.1109/TIT.2002.800479
