Estimating Source Term Parameters through Probabilistic Bayesian inference: An Approach based on an Adaptive Multiple Importance Sampling Algorithm

Harizo Rajaona; Patrick Armand; François Septier; Yves Delignon; Christophe Olry; Jacques Moussafir

Estimating Source Term Parameters through Probabilistic Bayesian inference: An Approach based on an Adaptive Multiple Importance Sampling Algorithm

Harizo Rajaona ^{1, 2, 3} Patrick Armand ³ François Septier ^{4, 1} Yves Delignon ^{1, 4} Christophe Olry ⁵ Jacques Moussafir ⁵

1 LAGIS-SI

LAGIS - Laboratoire d'Automatique, Génie Informatique et Signal

2 Ecole Centrale de Lille

3 DAM/DIF - DAM Île-de-France

4 Institut TELECOM/TELECOM Lille1

5 ARIA Technologies

Abstract : This paper presents an adaptive approach based on probabilistic Bayesian inference to estimate the parameters of an atmospheric pollution source term. After introducing the problem and assessing the computational framework, we present an Importance Sampling based algorithm called Adaptive Multiple Importance Sampling (AMIS). It performs an efficient calculation of the source parameter posterior distribution by iteratively upgrading the proposal's parameters and recycling all generations of weighted samples, thus allowing a faster convergence and reducing the number of necessary iterations. We highlight the results of the AMIS by comparing it to a MCMC estimation in a simple example.

Document type :

Conference papers

Domain :

Engineering Sciences [physics] / Signal and Image processing

Computer Science [cs] / Signal and Image Processing

Complete list of metadatas

https://hal-imt.archives-ouvertes.fr/hal-01064661
Contributor : François Septier <>
Submitted on : Tuesday, September 16, 2014 - 5:40:23 PM
Last modification on : Thursday, October 17, 2019 - 12:35:47 PM

Estimating Source Term Parameters through Probabilistic Bayesian inference: An Approach based on an Adaptive Multiple Importance Sampling Algorithm

Identifiers

Collections

Citation

Export

Metrics

289

Contact

Estimating Source Term Parameters through Probabilistic Bayesian inference: An Approach based on an Adaptive Multiple Importance Sampling Algorithm

Identifiers

Collections

Citation

Export

Share

Metrics

289

Contact