Expectation-Propagation algorithms for linear regression with Poisson noise: application to photon-limited spectral unmixing

Yoann Altmann; Alessandro Perelli; Mike E. Davies

doi:10.1109/ICASSP.2019.8682479

Expectation-Propagation algorithms for linear regression with Poisson noise: application to photon-limited spectral unmixing

Yoann Altmann, Alessandro Perelli, Mike E. Davies

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

7 Citations (Scopus)

94 Downloads (Pure)

Abstract

This paper discusses Expectation-Propagation (EP) methods for approximate Bayesian inference in the context of linear regression with Poisson noise. We review two main factor graphs used for generalized linear models and discuss how different EP algorithms can be derived. The estimation per- formance based on EP approximations is compared to the per- formance using Monte Carlo sampling from the exact poste- rior distribution. In particular, we observe that using locally independent or isotropic approximate factors enables more robust and scalable algorithms while providing reliable pos- terior means and marginal variances.

Original language	English
Title of host publication	2019 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)
Publisher	IEEE
Pages	5067-5071
Number of pages	5
ISBN (Electronic)	9781479981311
DOIs	https://doi.org/10.1109/ICASSP.2019.8682479
Publication status	Published - 17 Apr 2019
Event	44th IEEE International Conference on Acoustics, Speech, and Signal Processing 2019 - Brighton, United Kingdom Duration: 12 May 2019 → 17 May 2019

Conference

Conference	44th IEEE International Conference on Acoustics, Speech, and Signal Processing 2019
Abbreviated title	ICASSP 2019
Country/Territory	United Kingdom
City	Brighton
Period	12/05/19 → 17/05/19

Keywords

Approximate Bayesian inference
Expectation-Propagation
Poisson noise
linear regression

ASJC Scopus subject areas

Software
Signal Processing
Electrical and Electronic Engineering

Access to Document

10.1109/ICASSP.2019.8682479

Download pdf
© 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Accepted author manuscript, 625 KB

Cite this

@inproceedings{37522d0c952e4c149076f4925a1f8b99,

title = "Expectation-Propagation algorithms for linear regression with Poisson noise: application to photon-limited spectral unmixing",

abstract = "This paper discusses Expectation-Propagation (EP) methods for approximate Bayesian inference in the context of linear regression with Poisson noise. We review two main factor graphs used for generalized linear models and discuss how different EP algorithms can be derived. The estimation per- formance based on EP approximations is compared to the per- formance using Monte Carlo sampling from the exact poste- rior distribution. In particular, we observe that using locally independent or isotropic approximate factors enables more robust and scalable algorithms while providing reliable pos- terior means and marginal variances.",

keywords = "Approximate Bayesian inference, Expectation-Propagation, Poisson noise, linear regression",

author = "Yoann Altmann and Alessandro Perelli and Davies, {Mike E.}",

year = "2019",

month = apr,

day = "17",

doi = "10.1109/ICASSP.2019.8682479",

language = "English",

pages = "5067--5071",

booktitle = "2019 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)",

publisher = "IEEE",

address = "United States",

note = "44th IEEE International Conference on Acoustics, Speech, and Signal Processing 2019, ICASSP 2019 ; Conference date: 12-05-2019 Through 17-05-2019",

}

Altmann, Y, Perelli, A & Davies, ME 2019, Expectation-Propagation algorithms for linear regression with Poisson noise: application to photon-limited spectral unmixing. in 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)., 8682479, IEEE, pp. 5067-5071, 44th IEEE International Conference on Acoustics, Speech, and Signal Processing 2019, Brighton, United Kingdom, 12/05/19. https://doi.org/10.1109/ICASSP.2019.8682479

Expectation-Propagation algorithms for linear regression with Poisson noise: application to photon-limited spectral unmixing. / Altmann, Yoann; Perelli, Alessandro; Davies, Mike E.
2019 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP). IEEE, 2019. p. 5067-5071 8682479.

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

TY - GEN

T1 - Expectation-Propagation algorithms for linear regression with Poisson noise: application to photon-limited spectral unmixing

AU - Altmann, Yoann

AU - Perelli, Alessandro

AU - Davies, Mike E.

PY - 2019/4/17

Y1 - 2019/4/17

N2 - This paper discusses Expectation-Propagation (EP) methods for approximate Bayesian inference in the context of linear regression with Poisson noise. We review two main factor graphs used for generalized linear models and discuss how different EP algorithms can be derived. The estimation per- formance based on EP approximations is compared to the per- formance using Monte Carlo sampling from the exact poste- rior distribution. In particular, we observe that using locally independent or isotropic approximate factors enables more robust and scalable algorithms while providing reliable pos- terior means and marginal variances.

AB - This paper discusses Expectation-Propagation (EP) methods for approximate Bayesian inference in the context of linear regression with Poisson noise. We review two main factor graphs used for generalized linear models and discuss how different EP algorithms can be derived. The estimation per- formance based on EP approximations is compared to the per- formance using Monte Carlo sampling from the exact poste- rior distribution. In particular, we observe that using locally independent or isotropic approximate factors enables more robust and scalable algorithms while providing reliable pos- terior means and marginal variances.

KW - Approximate Bayesian inference

KW - Expectation-Propagation

KW - Poisson noise

KW - linear regression

UR - http://www.scopus.com/inward/record.url?scp=85068957633&partnerID=8YFLogxK

U2 - 10.1109/ICASSP.2019.8682479

DO - 10.1109/ICASSP.2019.8682479

M3 - Conference contribution

SP - 5067

EP - 5071

BT - 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

PB - IEEE

T2 - 44th IEEE International Conference on Acoustics, Speech, and Signal Processing 2019

Y2 - 12 May 2019 through 17 May 2019

ER -

Expectation-Propagation algorithms for linear regression with Poisson noise: application to photon-limited spectral unmixing

Abstract

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this