Communication requirement for distributed statistical machine learning with application in waveform cognition

Husheng Li; Zhu Han

doi:10.1109/GlobalSIP.2014.7032335

Communication requirement for distributed statistical machine learning with application in waveform cognition

Husheng Li
, Zhu Han

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Distributed learning is an effective approach to mitigate the data communications in machine learning when the data is stored in a distributed manner, particularly in the era of big data. In the distributed learning procedure, learners can send intermediate computation results instead of raw data, thus reducing the communication cost. In this paper, the communication requirement for distributed learning is studied in the scenario of multiple data storage nodes having the capability of learning and a fusion center. Lower bounds for communications are derived based on VC-entropy of modeling in the machine learning. Numerical results are provided to show the communication requirement for typical learning problems.

Original language	English
Title of host publication	2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1311-1314
Number of pages	4
ISBN (Electronic)	9781479970889
DOIs	https://doi.org/10.1109/GlobalSIP.2014.7032335
Publication status	Published - 5 Feb 2014
Event	2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014 - Atlanta, United States Duration: 3 Dec 2014 → 5 Dec 2014

Publication series

Name	2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014

Conference

Conference	2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014
Country/Territory	United States
City	Atlanta
Period	3/12/14 → 5/12/14

Bibliographical note

Publisher Copyright:
© 2014 IEEE.

Access to Document

10.1109/GlobalSIP.2014.7032335

Cite this

Li, H., & Han, Z. (2014). Communication requirement for distributed statistical machine learning with application in waveform cognition. In 2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014 (pp. 1311-1314). Article 7032335 (2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/GlobalSIP.2014.7032335

Li, Husheng ; Han, Zhu. / Communication requirement for distributed statistical machine learning with application in waveform cognition. 2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014. Institute of Electrical and Electronics Engineers Inc., 2014. pp. 1311-1314 (2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014).

@inproceedings{7a3f1ca56c264bc98713ce507f7606e7,

title = "Communication requirement for distributed statistical machine learning with application in waveform cognition",

abstract = "Distributed learning is an effective approach to mitigate the data communications in machine learning when the data is stored in a distributed manner, particularly in the era of big data. In the distributed learning procedure, learners can send intermediate computation results instead of raw data, thus reducing the communication cost. In this paper, the communication requirement for distributed learning is studied in the scenario of multiple data storage nodes having the capability of learning and a fusion center. Lower bounds for communications are derived based on VC-entropy of modeling in the machine learning. Numerical results are provided to show the communication requirement for typical learning problems.",

author = "Husheng Li and Zhu Han",

note = "Publisher Copyright: {\textcopyright} 2014 IEEE.; 2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014 ; Conference date: 03-12-2014 Through 05-12-2014",

year = "2014",

month = feb,

day = "5",

doi = "10.1109/GlobalSIP.2014.7032335",

language = "English",

series = "2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1311--1314",

booktitle = "2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014",

address = "United States",

}

Li, H & Han, Z 2014, Communication requirement for distributed statistical machine learning with application in waveform cognition. in 2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014., 7032335, 2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014, Institute of Electrical and Electronics Engineers Inc., pp. 1311-1314, 2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014, Atlanta, United States, 3/12/14. https://doi.org/10.1109/GlobalSIP.2014.7032335

Communication requirement for distributed statistical machine learning with application in waveform cognition. / Li, Husheng; Han, Zhu.
2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014. Institute of Electrical and Electronics Engineers Inc., 2014. p. 1311-1314 7032335 (2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Communication requirement for distributed statistical machine learning with application in waveform cognition

AU - Li, Husheng

AU - Han, Zhu

PY - 2014/2/5

Y1 - 2014/2/5

N2 - Distributed learning is an effective approach to mitigate the data communications in machine learning when the data is stored in a distributed manner, particularly in the era of big data. In the distributed learning procedure, learners can send intermediate computation results instead of raw data, thus reducing the communication cost. In this paper, the communication requirement for distributed learning is studied in the scenario of multiple data storage nodes having the capability of learning and a fusion center. Lower bounds for communications are derived based on VC-entropy of modeling in the machine learning. Numerical results are provided to show the communication requirement for typical learning problems.

AB - Distributed learning is an effective approach to mitigate the data communications in machine learning when the data is stored in a distributed manner, particularly in the era of big data. In the distributed learning procedure, learners can send intermediate computation results instead of raw data, thus reducing the communication cost. In this paper, the communication requirement for distributed learning is studied in the scenario of multiple data storage nodes having the capability of learning and a fusion center. Lower bounds for communications are derived based on VC-entropy of modeling in the machine learning. Numerical results are provided to show the communication requirement for typical learning problems.

UR - https://www.scopus.com/pages/publications/84983103294

U2 - 10.1109/GlobalSIP.2014.7032335

DO - 10.1109/GlobalSIP.2014.7032335

M3 - Conference contribution

AN - SCOPUS:84983103294

T3 - 2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014

SP - 1311

EP - 1314

BT - 2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014

Y2 - 3 December 2014 through 5 December 2014

ER -

Li H, Han Z. Communication requirement for distributed statistical machine learning with application in waveform cognition. In 2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014. Institute of Electrical and Electronics Engineers Inc. 2014. p. 1311-1314. 7032335. (2014 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2014). doi: 10.1109/GlobalSIP.2014.7032335

Communication requirement for distributed statistical machine learning with application in waveform cognition

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this