Multiple importance sampling with overlapping sets of proposals

Victor Elvira; Luca Martino; David Luengo; Monica F. Bugallo

doi:10.1109/SSP.2016.7551744

Multiple importance sampling with overlapping sets of proposals

Victor Elvira
, Luca Martino
, David Luengo
, Monica F. Bugallo

Electrical Engineering

Universidad Carlos III de Madrid
Universidade de São Paulo
Technical University of Madrid

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

7 Scopus citations

Abstract

In this paper, we introduce multiple importance sampling (MIS) approaches with overlapping (i.e., non-disjoint) sets of proposals. We derive a novel weighting scheme, based on the deterministic mixture methodology, that leads to unbiased estimators. The proposed framework can be seen as a generalization of other well-known MIS algorithms available in the literature. Furthermore, it allows us to achieve any desired trade-off between the variance of the estimators and the computational complexity through the definition of the sets of proposals. Simulations using a bimodal target density show the good performance of the proposed approach.

Original language	English
Title of host publication	2016 19th IEEE Statistical Signal Processing Workshop, SSP 2016
Publisher	IEEE Computer Society
ISBN (Electronic)	9781467378024
DOIs	https://doi.org/10.1109/SSP.2016.7551744
State	Published - Aug 24 2016
Event	19th IEEE Statistical Signal Processing Workshop, SSP 2016 - Palma de Mallorca, Spain Duration: Jun 25 2016 → Jun 29 2016

Publication series

Name	IEEE Workshop on Statistical Signal Processing Proceedings
Volume	2016-August

Conference

Conference	19th IEEE Statistical Signal Processing Workshop, SSP 2016
Country/Territory	Spain
City	Palma de Mallorca
Period	06/25/16 → 06/29/16

Keywords

Bayesian inference
Monte Carlo methods
Multiple importance sampling
variance reduction

Access to Document

10.1109/SSP.2016.7551744

Cite this

@inproceedings{31460b02a7a242009066969f3c608074,

title = "Multiple importance sampling with overlapping sets of proposals",

abstract = "In this paper, we introduce multiple importance sampling (MIS) approaches with overlapping (i.e., non-disjoint) sets of proposals. We derive a novel weighting scheme, based on the deterministic mixture methodology, that leads to unbiased estimators. The proposed framework can be seen as a generalization of other well-known MIS algorithms available in the literature. Furthermore, it allows us to achieve any desired trade-off between the variance of the estimators and the computational complexity through the definition of the sets of proposals. Simulations using a bimodal target density show the good performance of the proposed approach.",

keywords = "Bayesian inference, Monte Carlo methods, Multiple importance sampling, variance reduction",

author = "Victor Elvira and Luca Martino and David Luengo and Bugallo, \{Monica F.\}",

note = "Publisher Copyright: {\textcopyright} 2016 IEEE.; 19th IEEE Statistical Signal Processing Workshop, SSP 2016 ; Conference date: 25-06-2016 Through 29-06-2016",

year = "2016",

month = aug,

day = "24",

doi = "10.1109/SSP.2016.7551744",

language = "English",

series = "IEEE Workshop on Statistical Signal Processing Proceedings",

publisher = "IEEE Computer Society",

booktitle = "2016 19th IEEE Statistical Signal Processing Workshop, SSP 2016",

}

Elvira, V, Martino, L, Luengo, D & Bugallo, MF 2016, Multiple importance sampling with overlapping sets of proposals. in 2016 19th IEEE Statistical Signal Processing Workshop, SSP 2016., 7551744, IEEE Workshop on Statistical Signal Processing Proceedings, vol. 2016-August, IEEE Computer Society, 19th IEEE Statistical Signal Processing Workshop, SSP 2016, Palma de Mallorca, Spain, 06/25/16. https://doi.org/10.1109/SSP.2016.7551744

Multiple importance sampling with overlapping sets of proposals. / Elvira, Victor; Martino, Luca; Luengo, David et al.
2016 19th IEEE Statistical Signal Processing Workshop, SSP 2016. IEEE Computer Society, 2016. 7551744 (IEEE Workshop on Statistical Signal Processing Proceedings; Vol. 2016-August).

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

TY - GEN

T1 - Multiple importance sampling with overlapping sets of proposals

AU - Elvira, Victor

AU - Martino, Luca

AU - Luengo, David

AU - Bugallo, Monica F.

PY - 2016/8/24

Y1 - 2016/8/24

N2 - In this paper, we introduce multiple importance sampling (MIS) approaches with overlapping (i.e., non-disjoint) sets of proposals. We derive a novel weighting scheme, based on the deterministic mixture methodology, that leads to unbiased estimators. The proposed framework can be seen as a generalization of other well-known MIS algorithms available in the literature. Furthermore, it allows us to achieve any desired trade-off between the variance of the estimators and the computational complexity through the definition of the sets of proposals. Simulations using a bimodal target density show the good performance of the proposed approach.

AB - In this paper, we introduce multiple importance sampling (MIS) approaches with overlapping (i.e., non-disjoint) sets of proposals. We derive a novel weighting scheme, based on the deterministic mixture methodology, that leads to unbiased estimators. The proposed framework can be seen as a generalization of other well-known MIS algorithms available in the literature. Furthermore, it allows us to achieve any desired trade-off between the variance of the estimators and the computational complexity through the definition of the sets of proposals. Simulations using a bimodal target density show the good performance of the proposed approach.

KW - Bayesian inference

KW - Monte Carlo methods

KW - Multiple importance sampling

KW - variance reduction

UR - https://www.scopus.com/pages/publications/84987851755

U2 - 10.1109/SSP.2016.7551744

DO - 10.1109/SSP.2016.7551744

M3 - Conference contribution

AN - SCOPUS:84987851755

T3 - IEEE Workshop on Statistical Signal Processing Proceedings

BT - 2016 19th IEEE Statistical Signal Processing Workshop, SSP 2016

PB - IEEE Computer Society

T2 - 19th IEEE Statistical Signal Processing Workshop, SSP 2016

Y2 - 25 June 2016 through 29 June 2016

ER -

Multiple importance sampling with overlapping sets of proposals

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this