From Rate Distortion Theory to Metric Mean Dimension: Variational Principle

Elon Lindenstrauss, Masaki Tsukamoto

Research output: Contribution to journalArticle

Abstract

The purpose of this paper is to point out a new connection between information theory and dynamical systems. In the information theory side, we consider rate distortion theory, which studies lossy data compression of stochastic processes under distortion constraints. In the dynamical systems side, we consider mean dimension theory, which studies how many parameters per iterate we need to describe a dynamical system. The main results are new variational principles connecting rate distortion function to metric mean dimension.

Original languageEnglish
Pages (from-to)3590-3609
Number of pages20
JournalIEEE Transactions on Information Theory
Volume64
Issue number5
DOIs
Publication statusPublished - May 2018
Externally publishedYes

Fingerprint

Electric distortion
Variational techniques
Signal distortion
information theory
Stochastic systems
Information theory
entropy
Image coding
Random processes
Dynamical systems
Orbits
Entropy
Data compression

All Science Journal Classification (ASJC) codes

  • Information Systems
  • Computer Science Applications
  • Library and Information Sciences

Cite this

From Rate Distortion Theory to Metric Mean Dimension : Variational Principle. / Lindenstrauss, Elon; Tsukamoto, Masaki.

In: IEEE Transactions on Information Theory, Vol. 64, No. 5, 05.2018, p. 3590-3609.

Research output: Contribution to journalArticle

@article{d64fe1a5f0eb43d497457b950d6b968a,
title = "From Rate Distortion Theory to Metric Mean Dimension: Variational Principle",
abstract = "The purpose of this paper is to point out a new connection between information theory and dynamical systems. In the information theory side, we consider rate distortion theory, which studies lossy data compression of stochastic processes under distortion constraints. In the dynamical systems side, we consider mean dimension theory, which studies how many parameters per iterate we need to describe a dynamical system. The main results are new variational principles connecting rate distortion function to metric mean dimension.",
author = "Elon Lindenstrauss and Masaki Tsukamoto",
year = "2018",
month = "5",
doi = "10.1109/TIT.2018.2806219",
language = "English",
volume = "64",
pages = "3590--3609",
journal = "IEEE Transactions on Information Theory",
issn = "0018-9448",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "5",

}

TY - JOUR

T1 - From Rate Distortion Theory to Metric Mean Dimension

T2 - Variational Principle

AU - Lindenstrauss, Elon

AU - Tsukamoto, Masaki

PY - 2018/5

Y1 - 2018/5

N2 - The purpose of this paper is to point out a new connection between information theory and dynamical systems. In the information theory side, we consider rate distortion theory, which studies lossy data compression of stochastic processes under distortion constraints. In the dynamical systems side, we consider mean dimension theory, which studies how many parameters per iterate we need to describe a dynamical system. The main results are new variational principles connecting rate distortion function to metric mean dimension.

AB - The purpose of this paper is to point out a new connection between information theory and dynamical systems. In the information theory side, we consider rate distortion theory, which studies lossy data compression of stochastic processes under distortion constraints. In the dynamical systems side, we consider mean dimension theory, which studies how many parameters per iterate we need to describe a dynamical system. The main results are new variational principles connecting rate distortion function to metric mean dimension.

UR - http://www.scopus.com/inward/record.url?scp=85042855898&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85042855898&partnerID=8YFLogxK

U2 - 10.1109/TIT.2018.2806219

DO - 10.1109/TIT.2018.2806219

M3 - Article

AN - SCOPUS:85042855898

VL - 64

SP - 3590

EP - 3609

JO - IEEE Transactions on Information Theory

JF - IEEE Transactions on Information Theory

SN - 0018-9448

IS - 5

ER -