Clustering large networks of parametric dynamic generative models

Yunwen Xu, Sanggyun Kim, Srinivasa M Salapaka, Carolyn L Beck, Todd P. Coleman

Research output: Contribution to journalConference article

Abstract

Analysis, prediction and control of parametric generative models for stochastic processes arise in numerous applications, such as in biology, telecommunications, geography, seismology and finance. In many of these applications, it is desirable to obtain an aggregated behavior from an underlying network of stochastic interactions. This paper focuses on the simplification of parametric models describing multiple stochastic processes, by aggregating the processes that have similar input-output behaviors in an ensemble. We propose a clustering-based method, which is general in the sense that the similarity metric upon which the aggregation relies can accommodate processes characterized by a variety of generative models. To illustrate our aggregation framework, we investigate an example system comprised of a set of point process models for earthquakes. Simulations are presented.

Original languageEnglish (US)
Article number6425894
Pages (from-to)5248-5253
Number of pages6
JournalProceedings of the IEEE Conference on Decision and Control
DOIs
StatePublished - Dec 1 2012
Event51st IEEE Conference on Decision and Control, CDC 2012 - Maui, HI, United States
Duration: Dec 10 2012Dec 13 2012

Fingerprint

Generative Models
Parametric Model
Stochastic Processes
Dynamic models
Aggregation
Dynamic Model
Clustering
Seismology
Geography
Point Process
Random processes
Earthquake
Telecommunications
Finance
Set of points
Process Model
Simplification
Biology
Ensemble
Agglomeration

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Modeling and Simulation
  • Control and Optimization

Cite this

Clustering large networks of parametric dynamic generative models. / Xu, Yunwen; Kim, Sanggyun; Salapaka, Srinivasa M; Beck, Carolyn L; Coleman, Todd P.

In: Proceedings of the IEEE Conference on Decision and Control, 01.12.2012, p. 5248-5253.

Research output: Contribution to journalConference article

@article{ca2de2be0f2b4bf0a9dd1230c36ed725,
title = "Clustering large networks of parametric dynamic generative models",
abstract = "Analysis, prediction and control of parametric generative models for stochastic processes arise in numerous applications, such as in biology, telecommunications, geography, seismology and finance. In many of these applications, it is desirable to obtain an aggregated behavior from an underlying network of stochastic interactions. This paper focuses on the simplification of parametric models describing multiple stochastic processes, by aggregating the processes that have similar input-output behaviors in an ensemble. We propose a clustering-based method, which is general in the sense that the similarity metric upon which the aggregation relies can accommodate processes characterized by a variety of generative models. To illustrate our aggregation framework, we investigate an example system comprised of a set of point process models for earthquakes. Simulations are presented.",
author = "Yunwen Xu and Sanggyun Kim and Salapaka, {Srinivasa M} and Beck, {Carolyn L} and Coleman, {Todd P.}",
year = "2012",
month = "12",
day = "1",
doi = "10.1109/CDC.2012.6425894",
language = "English (US)",
pages = "5248--5253",
journal = "Proceedings of the IEEE Conference on Decision and Control",
issn = "0191-2216",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - JOUR

T1 - Clustering large networks of parametric dynamic generative models

AU - Xu, Yunwen

AU - Kim, Sanggyun

AU - Salapaka, Srinivasa M

AU - Beck, Carolyn L

AU - Coleman, Todd P.

PY - 2012/12/1

Y1 - 2012/12/1

N2 - Analysis, prediction and control of parametric generative models for stochastic processes arise in numerous applications, such as in biology, telecommunications, geography, seismology and finance. In many of these applications, it is desirable to obtain an aggregated behavior from an underlying network of stochastic interactions. This paper focuses on the simplification of parametric models describing multiple stochastic processes, by aggregating the processes that have similar input-output behaviors in an ensemble. We propose a clustering-based method, which is general in the sense that the similarity metric upon which the aggregation relies can accommodate processes characterized by a variety of generative models. To illustrate our aggregation framework, we investigate an example system comprised of a set of point process models for earthquakes. Simulations are presented.

AB - Analysis, prediction and control of parametric generative models for stochastic processes arise in numerous applications, such as in biology, telecommunications, geography, seismology and finance. In many of these applications, it is desirable to obtain an aggregated behavior from an underlying network of stochastic interactions. This paper focuses on the simplification of parametric models describing multiple stochastic processes, by aggregating the processes that have similar input-output behaviors in an ensemble. We propose a clustering-based method, which is general in the sense that the similarity metric upon which the aggregation relies can accommodate processes characterized by a variety of generative models. To illustrate our aggregation framework, we investigate an example system comprised of a set of point process models for earthquakes. Simulations are presented.

UR - http://www.scopus.com/inward/record.url?scp=84874247327&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84874247327&partnerID=8YFLogxK

U2 - 10.1109/CDC.2012.6425894

DO - 10.1109/CDC.2012.6425894

M3 - Conference article

AN - SCOPUS:84874247327

SP - 5248

EP - 5253

JO - Proceedings of the IEEE Conference on Decision and Control

JF - Proceedings of the IEEE Conference on Decision and Control

SN - 0191-2216

M1 - 6425894

ER -