Title :
A universal model based on minimax average divergence
Author :
Lu, Cheng-Chang ; Dunham, J.G.
Author_Institution :
Dept. of Math. Sci., Kent State Univ., OH, USA
fDate :
1/1/1992 12:00:00 AM
Abstract :
Given a set of training samples, the commonly used approach to determine a universal model is accomplished by averaging the statistics over all training samples. It is suggested to use average divergence as a measurement for the effectiveness of a universal model and a minimax universal model that minimizes the maximum average divergence among all training samples is proposed. Efficient searching algorithms are developed and experimental results are presented
Keywords :
data compression; encoding; information theory; minimax techniques; data compression; information theory; minimax average divergence; searching algorithms; source coding; training samples; universal model; Context modeling; Data compression; Encoding; Entropy; Information theory; Minimax techniques; Performance analysis; Source coding; Statistics; Stochastic processes;
Journal_Title :
Information Theory, IEEE Transactions on
