Are global sufficient statistics always sufficient: The impact of quantization on decentralized data reduction

Shengyu Zhu, Ge Xu, Biao Chen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

The sufficiency principle is the guiding principle for data reduction for various statistical inference problems. There has been recent effort in developing the sufficiency principle for decentralized inference with a particular emphasis on studying the relationship between global sufficient statistics and local sufficient statistics. We consider in this paper the impact of quantization on decentralized data reduction. The central question we intend to ask is: if each node in a decentralized inference system has to summarize its data using a finite number of bits, is it still sufficient to implement data reduction using global sufficient statistics prior to quantization? We show that the answer is negative using a simple example and proceed to identify conditions when global sufficient statistics based data reduction is indeed optimal. They include the well known case when the data at decentralized nodes are conditionally independent as well as a class of problems with conditionally dependent data.

Original languageEnglish (US)
Title of host publicationConference Record - Asilomar Conference on Signals, Systems and Computers
PublisherIEEE Computer Society
Pages1090-1094
Number of pages5
ISBN (Print)9781479923908
DOIs
StatePublished - 2013
Event2013 47th Asilomar Conference on Signals, Systems and Computers - Pacific Grove, CA, United States
Duration: Nov 3 2013Nov 6 2013

Other

Other2013 47th Asilomar Conference on Signals, Systems and Computers
CountryUnited States
CityPacific Grove, CA
Period11/3/1311/6/13

Fingerprint

Data reduction
Statistics

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Signal Processing

Cite this

Zhu, S., Xu, G., & Chen, B. (2013). Are global sufficient statistics always sufficient: The impact of quantization on decentralized data reduction. In Conference Record - Asilomar Conference on Signals, Systems and Computers (pp. 1090-1094). [6810461] IEEE Computer Society. https://doi.org/10.1109/ACSSC.2013.6810461

Are global sufficient statistics always sufficient : The impact of quantization on decentralized data reduction. / Zhu, Shengyu; Xu, Ge; Chen, Biao.

Conference Record - Asilomar Conference on Signals, Systems and Computers. IEEE Computer Society, 2013. p. 1090-1094 6810461.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Zhu, S, Xu, G & Chen, B 2013, Are global sufficient statistics always sufficient: The impact of quantization on decentralized data reduction. in Conference Record - Asilomar Conference on Signals, Systems and Computers., 6810461, IEEE Computer Society, pp. 1090-1094, 2013 47th Asilomar Conference on Signals, Systems and Computers, Pacific Grove, CA, United States, 11/3/13. https://doi.org/10.1109/ACSSC.2013.6810461
Zhu S, Xu G, Chen B. Are global sufficient statistics always sufficient: The impact of quantization on decentralized data reduction. In Conference Record - Asilomar Conference on Signals, Systems and Computers. IEEE Computer Society. 2013. p. 1090-1094. 6810461 https://doi.org/10.1109/ACSSC.2013.6810461
Zhu, Shengyu ; Xu, Ge ; Chen, Biao. / Are global sufficient statistics always sufficient : The impact of quantization on decentralized data reduction. Conference Record - Asilomar Conference on Signals, Systems and Computers. IEEE Computer Society, 2013. pp. 1090-1094
@inproceedings{db16355c521e4b6791cc1dfce43d399a,
title = "Are global sufficient statistics always sufficient: The impact of quantization on decentralized data reduction",
abstract = "The sufficiency principle is the guiding principle for data reduction for various statistical inference problems. There has been recent effort in developing the sufficiency principle for decentralized inference with a particular emphasis on studying the relationship between global sufficient statistics and local sufficient statistics. We consider in this paper the impact of quantization on decentralized data reduction. The central question we intend to ask is: if each node in a decentralized inference system has to summarize its data using a finite number of bits, is it still sufficient to implement data reduction using global sufficient statistics prior to quantization? We show that the answer is negative using a simple example and proceed to identify conditions when global sufficient statistics based data reduction is indeed optimal. They include the well known case when the data at decentralized nodes are conditionally independent as well as a class of problems with conditionally dependent data.",
author = "Shengyu Zhu and Ge Xu and Biao Chen",
year = "2013",
doi = "10.1109/ACSSC.2013.6810461",
language = "English (US)",
isbn = "9781479923908",
pages = "1090--1094",
booktitle = "Conference Record - Asilomar Conference on Signals, Systems and Computers",
publisher = "IEEE Computer Society",
address = "United States",

}

TY - GEN

T1 - Are global sufficient statistics always sufficient

T2 - The impact of quantization on decentralized data reduction

AU - Zhu, Shengyu

AU - Xu, Ge

AU - Chen, Biao

PY - 2013

Y1 - 2013

N2 - The sufficiency principle is the guiding principle for data reduction for various statistical inference problems. There has been recent effort in developing the sufficiency principle for decentralized inference with a particular emphasis on studying the relationship between global sufficient statistics and local sufficient statistics. We consider in this paper the impact of quantization on decentralized data reduction. The central question we intend to ask is: if each node in a decentralized inference system has to summarize its data using a finite number of bits, is it still sufficient to implement data reduction using global sufficient statistics prior to quantization? We show that the answer is negative using a simple example and proceed to identify conditions when global sufficient statistics based data reduction is indeed optimal. They include the well known case when the data at decentralized nodes are conditionally independent as well as a class of problems with conditionally dependent data.

AB - The sufficiency principle is the guiding principle for data reduction for various statistical inference problems. There has been recent effort in developing the sufficiency principle for decentralized inference with a particular emphasis on studying the relationship between global sufficient statistics and local sufficient statistics. We consider in this paper the impact of quantization on decentralized data reduction. The central question we intend to ask is: if each node in a decentralized inference system has to summarize its data using a finite number of bits, is it still sufficient to implement data reduction using global sufficient statistics prior to quantization? We show that the answer is negative using a simple example and proceed to identify conditions when global sufficient statistics based data reduction is indeed optimal. They include the well known case when the data at decentralized nodes are conditionally independent as well as a class of problems with conditionally dependent data.

UR - http://www.scopus.com/inward/record.url?scp=84901248096&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84901248096&partnerID=8YFLogxK

U2 - 10.1109/ACSSC.2013.6810461

DO - 10.1109/ACSSC.2013.6810461

M3 - Conference contribution

AN - SCOPUS:84901248096

SN - 9781479923908

SP - 1090

EP - 1094

BT - Conference Record - Asilomar Conference on Signals, Systems and Computers

PB - IEEE Computer Society

ER -