1. Rényi Entropy Power and Normal Transport
- Author
-
Rioul, Olivier, Communications Numériques (COMNUM), Laboratoire Traitement et Communication de l'Information (LTCI), Institut Mines-Télécom [Paris] (IMT)-Télécom Paris-Institut Mines-Télécom [Paris] (IMT)-Télécom Paris, Département Communications & Electronique (COMELEC), Télécom ParisTech, IEICE, and Rioul, Olivier
- Subjects
[MATH.MATH-PR] Mathematics [math]/Probability [math.PR] ,[INFO.INFO-TS] Computer Science [cs]/Signal and Image Processing ,[MATH.MATH-FA] Mathematics [math]/Functional Analysis [math.FA] ,[MATH.MATH-IT]Mathematics [math]/Information Theory [math.IT] ,[MATH.MATH-GM] Mathematics [math]/General Mathematics [math.GM] ,[MATH.MATH-CA]Mathematics [math]/Classical Analysis and ODEs [math.CA] ,[INFO.INFO-DM]Computer Science [cs]/Discrete Mathematics [cs.DM] ,[MATH.MATH-FA]Mathematics [math]/Functional Analysis [math.FA] ,[MATH.MATH-CA] Mathematics [math]/Classical Analysis and ODEs [math.CA] ,[MATH.MATH-PR]Mathematics [math]/Probability [math.PR] ,[MATH.MATH-IT] Mathematics [math]/Information Theory [math.IT] ,[INFO.INFO-DM] Computer Science [cs]/Discrete Mathematics [cs.DM] ,[INFO.INFO-CR]Computer Science [cs]/Cryptography and Security [cs.CR] ,[MATH.MATH-GM]Mathematics [math]/General Mathematics [math.GM] ,[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing ,[INFO.INFO-IT]Computer Science [cs]/Information Theory [cs.IT] ,[MATH.MATH-ST]Mathematics [math]/Statistics [math.ST] ,[INFO.INFO-IT] Computer Science [cs]/Information Theory [cs.IT] ,[INFO.INFO-HC]Computer Science [cs]/Human-Computer Interaction [cs.HC] ,[INFO.INFO-HC] Computer Science [cs]/Human-Computer Interaction [cs.HC] ,[MATH.MATH-ST] Mathematics [math]/Statistics [math.ST] ,[SPI.SIGNAL]Engineering Sciences [physics]/Signal and Image processing ,[SPI.SIGNAL] Engineering Sciences [physics]/Signal and Image processing ,[INFO.INFO-CR] Computer Science [cs]/Cryptography and Security [cs.CR] - Abstract
International audience; A framework for deriving Rényi entropy-power inequalities (REPIs) is presented that uses linearization and an inequality of Dembo, Cover, and Thomas. Simple arguments are given to recover the previously known Rényi EPIs and derive new ones, by unifying a multiplicative form with constant c and a modification with exponent α of previous works. An information-theoretic proof of the Dembo-Cover-Thomas inequality-equivalent to Young's convolutional inequality with optimal constants-is provided, based on properties of Rényi conditional and relative entropies and using transportation arguments from Gaussian densities. For log-concave densities, a transportation proof of a sharp varentropy bound is presented. This work was partially presented at the 2019 Information Theory and Applications Workshop, San Diego, CA.
- Published
- 2020