A Bayesian parameter learning procedure for nonlinear dynamical systems via the ensemble Kalman filter

Muhammad Javvad Ur Rehman; Sarat C. Dass; Vijanth S. Asirvadam

doi:10.1109/CSPA.2018.8368705

A Bayesian parameter learning procedure for nonlinear dynamical systems via the ensemble Kalman filter

Muhammad Javvad Ur Rehman, Sarat C. Dass, Vijanth S. Asirvadam

School of Mathematical & Computer Sciences

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

2 Citations (Scopus)

Abstract

Dynamical systems are a natural and convenient way to model the evolution of processes observed in practice. When uncertainty is considered and incorporated, these system become known as stochastic dynamical systems. Based on observations made from stochastic dynamical systems, we consider the issue of parameter learning, and a related state estimation problem. We develop a Markov Chain Monte Carlo (MCMC) algorithm, which is an iterative method, for parameter inference. Within the parameter learning steps, the MCMC algorithm requires to perform state estimation for which the target distribution is constructed by using the Ensemble Kalman filter (EnKF). The methodology is illustrated using two examples of nonlinear stochastic dynamical systems.

Original language	English
Title of host publication	2018 IEEE 14th International Colloquium on Signal Processing & Its Applications (CSPA)
Publisher	IEEE
Pages	161-166
Number of pages	6
ISBN (Electronic)	9781538603895
DOIs	https://doi.org/10.1109/CSPA.2018.8368705
Publication status	Published - 31 May 2018
Event	14th IEEE International Colloquium on Signal Processing and its Application 2018 - Batu Feringghi, Penang, Malaysia Duration: 9 Mar 2018 → 10 Mar 2018

Conference

Conference	14th IEEE International Colloquium on Signal Processing and its Application 2018
Abbreviated title	CSPA 2018
Country/Territory	Malaysia
City	Penang
Period	9/03/18 → 10/03/18

ASJC Scopus subject areas

Computer Science Applications
Computer Vision and Pattern Recognition
Signal Processing

Access to Document

10.1109/CSPA.2018.8368705

Cite this

@inproceedings{8d5e2ca873b443838347e7160e47ee72,

title = "A Bayesian parameter learning procedure for nonlinear dynamical systems via the ensemble Kalman filter",

abstract = "Dynamical systems are a natural and convenient way to model the evolution of processes observed in practice. When uncertainty is considered and incorporated, these system become known as stochastic dynamical systems. Based on observations made from stochastic dynamical systems, we consider the issue of parameter learning, and a related state estimation problem. We develop a Markov Chain Monte Carlo (MCMC) algorithm, which is an iterative method, for parameter inference. Within the parameter learning steps, the MCMC algorithm requires to perform state estimation for which the target distribution is constructed by using the Ensemble Kalman filter (EnKF). The methodology is illustrated using two examples of nonlinear stochastic dynamical systems.",

author = "Rehman, {Muhammad Javvad Ur} and Dass, {Sarat C.} and Asirvadam, {Vijanth S.}",

year = "2018",

month = may,

day = "31",

doi = "10.1109/CSPA.2018.8368705",

language = "English",

pages = "161--166",

booktitle = "2018 IEEE 14th International Colloquium on Signal Processing & Its Applications (CSPA)",

publisher = "IEEE",

address = "United States",

note = "14th IEEE International Colloquium on Signal Processing and its Application 2018, CSPA 2018 ; Conference date: 09-03-2018 Through 10-03-2018",

}

Rehman, MJU, Dass, SC & Asirvadam, VS 2018, A Bayesian parameter learning procedure for nonlinear dynamical systems via the ensemble Kalman filter. in 2018 IEEE 14th International Colloquium on Signal Processing & Its Applications (CSPA). IEEE, pp. 161-166, 14th IEEE International Colloquium on Signal Processing and its Application 2018, Penang, Malaysia, 9/03/18. https://doi.org/10.1109/CSPA.2018.8368705

A Bayesian parameter learning procedure for nonlinear dynamical systems via the ensemble Kalman filter. / Rehman, Muhammad Javvad Ur; Dass, Sarat C.; Asirvadam, Vijanth S.
2018 IEEE 14th International Colloquium on Signal Processing & Its Applications (CSPA). IEEE, 2018. p. 161-166.

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

TY - GEN

T1 - A Bayesian parameter learning procedure for nonlinear dynamical systems via the ensemble Kalman filter

AU - Rehman, Muhammad Javvad Ur

AU - Dass, Sarat C.

AU - Asirvadam, Vijanth S.

PY - 2018/5/31

Y1 - 2018/5/31

N2 - Dynamical systems are a natural and convenient way to model the evolution of processes observed in practice. When uncertainty is considered and incorporated, these system become known as stochastic dynamical systems. Based on observations made from stochastic dynamical systems, we consider the issue of parameter learning, and a related state estimation problem. We develop a Markov Chain Monte Carlo (MCMC) algorithm, which is an iterative method, for parameter inference. Within the parameter learning steps, the MCMC algorithm requires to perform state estimation for which the target distribution is constructed by using the Ensemble Kalman filter (EnKF). The methodology is illustrated using two examples of nonlinear stochastic dynamical systems.

AB - Dynamical systems are a natural and convenient way to model the evolution of processes observed in practice. When uncertainty is considered and incorporated, these system become known as stochastic dynamical systems. Based on observations made from stochastic dynamical systems, we consider the issue of parameter learning, and a related state estimation problem. We develop a Markov Chain Monte Carlo (MCMC) algorithm, which is an iterative method, for parameter inference. Within the parameter learning steps, the MCMC algorithm requires to perform state estimation for which the target distribution is constructed by using the Ensemble Kalman filter (EnKF). The methodology is illustrated using two examples of nonlinear stochastic dynamical systems.

UR - http://www.scopus.com/inward/record.url?scp=85048821948&partnerID=8YFLogxK

U2 - 10.1109/CSPA.2018.8368705

DO - 10.1109/CSPA.2018.8368705

M3 - Conference contribution

AN - SCOPUS:85048821948

SP - 161

EP - 166

BT - 2018 IEEE 14th International Colloquium on Signal Processing & Its Applications (CSPA)

PB - IEEE

T2 - 14th IEEE International Colloquium on Signal Processing and its Application 2018

Y2 - 9 March 2018 through 10 March 2018

ER -

A Bayesian parameter learning procedure for nonlinear dynamical systems via the ensemble Kalman filter

Abstract

Conference

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this