Measures of Serial Data Compressibility by Neural Network Predictors

James P. Coughlin; R. H. Baran; Hanseok Ko

doi:10.1109/IJCNN.1992.287096

Measures of Serial Data Compressibility by Neural Network Predictors

James P. Coughlin, R. H. Baran, Hanseok Ko

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

Abstract

A time series or univariate random process is compressible if it is predictable. Experiments with a variety of processes readily show that adaptive neural networks are at least as effective as their linear counterparts in one-step-ahead prediction. We explore the relationship between the predictive accuracy attained by the network, in the long run, and the closeness with which it can fit (and overfit) small segments of the same series in the course of many passes through the same data. Our findings suggest that the predictability of a process can be estimated by measuring the ease with which its increments can be overfitted.

Original language	English
Title of host publication	Proceedings - 1992 International Joint Conference on Neural Networks, IJCNN 1992
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	755-760
Number of pages	6
ISBN (Electronic)	0780305590
DOIs	https://doi.org/10.1109/IJCNN.1992.287096
Publication status	Published - 1992
Externally published	Yes
Event	1992 International Joint Conference on Neural Networks, IJCNN 1992 - Baltimore, United States Duration: 1992 Jun 7 → 1992 Jun 11

Publication series

Name	Proceedings of the International Joint Conference on Neural Networks
Volume	1

Conference

Conference	1992 International Joint Conference on Neural Networks, IJCNN 1992
Country/Territory	United States
City	Baltimore
Period	92/6/7 → 92/6/11

Bibliographical note

Publisher Copyright:
© 1992 IEEE.

ASJC Scopus subject areas

Software
Artificial Intelligence

Access to Document

10.1109/IJCNN.1992.287096

Cite this

Coughlin, J. P., Baran, R. H., & Ko, H. (1992). Measures of Serial Data Compressibility by Neural Network Predictors. In Proceedings - 1992 International Joint Conference on Neural Networks, IJCNN 1992 (pp. 755-760). (Proceedings of the International Joint Conference on Neural Networks; Vol. 1). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IJCNN.1992.287096

Measures of Serial Data Compressibility by Neural Network Predictors. / Coughlin, James P.; Baran, R. H.; Ko, Hanseok.
Proceedings - 1992 International Joint Conference on Neural Networks, IJCNN 1992. Institute of Electrical and Electronics Engineers Inc., 1992. p. 755-760 (Proceedings of the International Joint Conference on Neural Networks; Vol. 1).

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

Coughlin, JP, Baran, RH & Ko, H 1992, Measures of Serial Data Compressibility by Neural Network Predictors. in Proceedings - 1992 International Joint Conference on Neural Networks, IJCNN 1992. Proceedings of the International Joint Conference on Neural Networks, vol. 1, Institute of Electrical and Electronics Engineers Inc., pp. 755-760, 1992 International Joint Conference on Neural Networks, IJCNN 1992, Baltimore, United States, 92/6/7. https://doi.org/10.1109/IJCNN.1992.287096

@inproceedings{b24b81afd5994409871aac564178f4f7,

title = "Measures of Serial Data Compressibility by Neural Network Predictors",

abstract = "A time series or univariate random process is compressible if it is predictable. Experiments with a variety of processes readily show that adaptive neural networks are at least as effective as their linear counterparts in one-step-ahead prediction. We explore the relationship between the predictive accuracy attained by the network, in the long run, and the closeness with which it can fit (and overfit) small segments of the same series in the course of many passes through the same data. Our findings suggest that the predictability of a process can be estimated by measuring the ease with which its increments can be overfitted.",

author = "Coughlin, {James P.} and Baran, {R. H.} and Hanseok Ko",

note = "Publisher Copyright: {\textcopyright} 1992 IEEE.; 1992 International Joint Conference on Neural Networks, IJCNN 1992 ; Conference date: 07-06-1992 Through 11-06-1992",

year = "1992",

doi = "10.1109/IJCNN.1992.287096",

language = "English",

series = "Proceedings of the International Joint Conference on Neural Networks",

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

pages = "755--760",

booktitle = "Proceedings - 1992 International Joint Conference on Neural Networks, IJCNN 1992",

}

TY - GEN

T1 - Measures of Serial Data Compressibility by Neural Network Predictors

AU - Coughlin, James P.

AU - Baran, R. H.

AU - Ko, Hanseok

PY - 1992

Y1 - 1992

N2 - A time series or univariate random process is compressible if it is predictable. Experiments with a variety of processes readily show that adaptive neural networks are at least as effective as their linear counterparts in one-step-ahead prediction. We explore the relationship between the predictive accuracy attained by the network, in the long run, and the closeness with which it can fit (and overfit) small segments of the same series in the course of many passes through the same data. Our findings suggest that the predictability of a process can be estimated by measuring the ease with which its increments can be overfitted.

AB - A time series or univariate random process is compressible if it is predictable. Experiments with a variety of processes readily show that adaptive neural networks are at least as effective as their linear counterparts in one-step-ahead prediction. We explore the relationship between the predictive accuracy attained by the network, in the long run, and the closeness with which it can fit (and overfit) small segments of the same series in the course of many passes through the same data. Our findings suggest that the predictability of a process can be estimated by measuring the ease with which its increments can be overfitted.

UR - http://www.scopus.com/inward/record.url?scp=85132006007&partnerID=8YFLogxK

U2 - 10.1109/IJCNN.1992.287096

DO - 10.1109/IJCNN.1992.287096

M3 - Conference contribution

AN - SCOPUS:85132006007

T3 - Proceedings of the International Joint Conference on Neural Networks

SP - 755

EP - 760

BT - Proceedings - 1992 International Joint Conference on Neural Networks, IJCNN 1992

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 1992 International Joint Conference on Neural Networks, IJCNN 1992

Y2 - 7 June 1992 through 11 June 1992

ER -

Measures of Serial Data Compressibility by Neural Network Predictors

Abstract

Publication series

Conference

Bibliographical note

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this