Title :
Convergence of best φ-entropy estimates
Author :
Teboulle, Marc ; Vajda, Igor
Author_Institution :
Dept. of Math. & Stat., Maryland Univ., Baltimore, MD, USA
fDate :
1/1/1993 12:00:00 AM
Abstract :
Minimization problems involving φ-entropy functionals (a generalization of Boltzmann-Shannon entropy) are studied over a given set A and a sequence of sets An and the properties of their optimal solutions xφ, x n. Under certain conditions on the objective functional and the sets A and An, it is proven that as n increases to infinity, the optimal solution xn converges in L1 norm to the test φ-entropy estimate xφ
Keywords :
convergence; entropy; estimation theory; information theory; minimisation; Boltzmann-Shannon entropy; best φ-entropy estimates; convergence; information theory; minimisation; objective functional; Arithmetic; Automation; Convergence; Entropy; Functional analysis; H infinity control; Information theory; Mathematics; Statistics; Testing; White noise;
Journal_Title :
Information Theory, IEEE Transactions on
