Compression by replication

Roberto C. Alamino*, Juan P. Neirotti, David Saad

*Corresponding author for this work

Research output: Contribution to journalArticle

Abstract

A recently introduced inference method based on system replication and an online message passing algorithm is employed to complete a previously suggested compression scheme based on a nonlinear perceptron. The algorithm is shown to approach the information theoretical bounds for compression as the number of replicated systems increases, offering superior performance compared to basic message passing algorithms. In addition, the suggested method does not require fine-tuning of parameters or other complementing heuristic techniques, such as the introduction of inertia terms, to improve convergence rates to nontrivial results.

Original languageEnglish
Article number033301
Number of pages7
JournalPhysical Review E
Volume89
Issue number3
DOIs
Publication statusPublished - 4 Mar 2014

Fingerprint

Message-passing Algorithms
Replication
Compression
messages
Online Algorithms
Perceptron
Inertia
Convergence Rate
Tuning
self organizing systems
Heuristics
inference
inertia
Term
tuning

Cite this

@article{f299eb71aba64c198b646dbe421be7d0,
title = "Compression by replication",
abstract = "A recently introduced inference method based on system replication and an online message passing algorithm is employed to complete a previously suggested compression scheme based on a nonlinear perceptron. The algorithm is shown to approach the information theoretical bounds for compression as the number of replicated systems increases, offering superior performance compared to basic message passing algorithms. In addition, the suggested method does not require fine-tuning of parameters or other complementing heuristic techniques, such as the introduction of inertia terms, to improve convergence rates to nontrivial results.",
author = "Alamino, {Roberto C.} and Neirotti, {Juan P.} and David Saad",
year = "2014",
month = "3",
day = "4",
doi = "10.1103/PhysRevE.89.033301",
language = "English",
volume = "89",
journal = "Physical Review E",
issn = "1539-3755",
publisher = "American Physical Society",
number = "3",

}

Compression by replication. / Alamino, Roberto C.; Neirotti, Juan P.; Saad, David.

In: Physical Review E, Vol. 89, No. 3, 033301, 04.03.2014.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Compression by replication

AU - Alamino, Roberto C.

AU - Neirotti, Juan P.

AU - Saad, David

PY - 2014/3/4

Y1 - 2014/3/4

N2 - A recently introduced inference method based on system replication and an online message passing algorithm is employed to complete a previously suggested compression scheme based on a nonlinear perceptron. The algorithm is shown to approach the information theoretical bounds for compression as the number of replicated systems increases, offering superior performance compared to basic message passing algorithms. In addition, the suggested method does not require fine-tuning of parameters or other complementing heuristic techniques, such as the introduction of inertia terms, to improve convergence rates to nontrivial results.

AB - A recently introduced inference method based on system replication and an online message passing algorithm is employed to complete a previously suggested compression scheme based on a nonlinear perceptron. The algorithm is shown to approach the information theoretical bounds for compression as the number of replicated systems increases, offering superior performance compared to basic message passing algorithms. In addition, the suggested method does not require fine-tuning of parameters or other complementing heuristic techniques, such as the introduction of inertia terms, to improve convergence rates to nontrivial results.

UR - http://www.scopus.com/inward/record.url?scp=84898928957&partnerID=8YFLogxK

U2 - 10.1103/PhysRevE.89.033301

DO - 10.1103/PhysRevE.89.033301

M3 - Article

AN - SCOPUS:84898928957

VL - 89

JO - Physical Review E

JF - Physical Review E

SN - 1539-3755

IS - 3

M1 - 033301

ER -