Title :
A universal prediction lemma and applications to universal data compression and prediction
Author_Institution :
Dept. of Electr. Eng., Technion-Israel Inst. of Technol., Haifa, Israel
Abstract :
We consider finite-alphabet sequences which are emitted by a stationary source with unknown statistics. We treat the optimization problem by deriving performance bounds for a restricted class of empirical conditional distributions (predictors)
Keywords :
data compression; optimisation; prediction theory; probability; sequences; source coding; empirical conditional distributions; finite-alphabet sequences; optimization problem; performance bounds; predictors; stationary source; universal data compression; universal prediction lemma; unknown statistics; Data compression; Entropy; Frequency measurement; Jacobian matrices; Minimax techniques; Probability; Random variables; Statistics;
Conference_Titel :
Information Theory, 2000. Proceedings. IEEE International Symposium on
Conference_Location :
Sorrento
Print_ISBN :
0-7803-5857-0
DOI :
10.1109/ISIT.2000.866359
