Joint universal lossy coding and identification of stationary mixing sources - Computer Science Information TheoryReportar como inadecuado




Joint universal lossy coding and identification of stationary mixing sources - Computer Science Information Theory - Descarga este documento en PDF. Documentación en PDF para descargar gratis. Disponible también para leer online.

Abstract: The problem of joint universal source coding and modeling, treated in thecontext of lossless codes by Rissanen, was recently generalized to fixed-ratelossy coding of finitely parametrized continuous-alphabet i.i.d. sources. Weextend these results to variable-rate lossy block coding of stationary ergodicsources and show that, for bounded metric distortion measures, any finitelyparametrized family of stationary sources satisfying suitable mixing,smoothness and Vapnik-Chervonenkis learnability conditions admits universalschemes for joint lossy source coding and identification. We also give severalexplicit examples of parametric sources satisfying the regularity conditions.



Autor: Maxim Raginsky

Fuente: https://arxiv.org/







Documentos relacionados