Skip to main content
Top

2021 | OriginalPaper | Chapter

Robust Linear Regression and Anomaly Detection in the Presence of Poisson Noise Using Expectation-Propagation

Authors : Yoann Altmann, Dan Yao, Stephen McLaughlin, Mike E. Davies

Published in: Advances in Condition Monitoring and Structural Health Monitoring

Publisher: Springer Singapore

Activate our intelligent search to find suitable subject content or patents.

search-config
loading …

Abstract

This paper presents a family of approximate Bayesian methods for joint anomaly detection and linear regression in the presence of non-Gaussian noise. Robust anomaly detection using non-convex sparsity-promoting regularization terms is generally challenging, in particular when additional uncertainty measures about the estimation process are needed, e.g., posterior probabilities of anomaly presence. The problem becomes even more challenging in the presence of non-Gaussian, (e.g., Poisson distributed), additional constraints on the regression coefficients (e.g., positivity) and when the anomalies present complex structures (e.g., structured sparsity). Uncertainty quantification is classically addressed using Bayesian methods. Specifically, Monte Carlo methods are the preferred tools to handle complex models. Unfortunately, such simulation methods suffer from a significant computational cost and are thus not scalable for fast inference in high dimensional problems. In this paper, we thus propose fast alternatives based on Expectation-Propagation (EP) methods, which aim at approximating complex distributions by more tractable models to simplify the inference process. The main problem addressed in this paper is linear regression and (sparse) anomaly detection in the presence of noisy measurements. The aim of this paper is to demonstrate the potential benefits and assess the performance of such EP-based methods. The results obtained illustrate that approximate methods can provide satisfactory results with a reasonable computational cost. It is important to note that the proposed methods are sufficiently generic to be used in other applications involving condition monitoring.

Dont have a licence yet? Then find out more about our products and how to get one now:

Springer Professional "Wirtschaft+Technik"

Online-Abonnement

Mit Springer Professional "Wirtschaft+Technik" erhalten Sie Zugriff auf:

  • über 102.000 Bücher
  • über 537 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Nachhaltigkeit
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Maschinenbau + Werkstoffe
  • Versicherung + Risiko

Jetzt Wissensvorsprung sichern!

Springer Professional "Technik"

Online-Abonnement

Mit Springer Professional "Technik" erhalten Sie Zugriff auf:

  • über 67.000 Bücher
  • über 390 Zeitschriften

aus folgenden Fachgebieten:

  • Automobil + Motoren
  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Elektrotechnik + Elektronik
  • Energie + Nachhaltigkeit
  • Maschinenbau + Werkstoffe




 

Jetzt Wissensvorsprung sichern!

Springer Professional "Wirtschaft"

Online-Abonnement

Mit Springer Professional "Wirtschaft" erhalten Sie Zugriff auf:

  • über 67.000 Bücher
  • über 340 Zeitschriften

aus folgenden Fachgebieten:

  • Bauwesen + Immobilien
  • Business IT + Informatik
  • Finance + Banking
  • Management + Führung
  • Marketing + Vertrieb
  • Versicherung + Risiko




Jetzt Wissensvorsprung sichern!

Footnotes
1
Note that VB methods are also based on KL-divergence minimization, but they use the direct KL-divergence \(KL\left( Q(\varvec{u},{{\varvec{x}}},\varvec{r},\tilde{{\varvec{z}}}) || f(\varvec{u},{{\varvec{x}}},\varvec{r},{\varvec{z}},{\varvec{y}})\right) \).
 
Literature
1.
go back to reference Newstadt GE, III AH, Simmons J (2014) Robust spectral unmixing for anomaly detection. In: Proceeding IEEE-SP workshop statistical and saignal processing, Gold Coast, Australia Newstadt GE, III AH, Simmons J (2014) Robust spectral unmixing for anomaly detection. In: Proceeding IEEE-SP workshop statistical and saignal processing, Gold Coast, Australia
2.
go back to reference Altmann Y, Maccarone A, McCarthy A, Newstadt G, Buller GS, McLaughlin S, Hero A (2017) Robust spectral unmixing of sparse multispectral lidar waveforms using gamma Markov random fields. IEEE Trans Comput Imaging 3(4):658–670MathSciNetCrossRef Altmann Y, Maccarone A, McCarthy A, Newstadt G, Buller GS, McLaughlin S, Hero A (2017) Robust spectral unmixing of sparse multispectral lidar waveforms using gamma Markov random fields. IEEE Trans Comput Imaging 3(4):658–670MathSciNetCrossRef
3.
go back to reference Tachella J, ltmann Y, Pereyra M, Tourneret J-Y (2018) Bayesian restoration of high-dimensional photon-starved images. In: Proceeding European signal processing conference (EUSIPCO), Rome, Italy Tachella J, ltmann Y, Pereyra M, Tourneret J-Y (2018) Bayesian restoration of high-dimensional photon-starved images. In: Proceeding European signal processing conference (EUSIPCO), Rome, Italy
4.
go back to reference Blei DM, Kucukelbir A, McAuliffe JD (2017) Variational inference: a review for statisticians. J Am Stat Assoc 112(518):859–877MathSciNetCrossRef Blei DM, Kucukelbir A, McAuliffe JD (2017) Variational inference: a review for statisticians. J Am Stat Assoc 112(518):859–877MathSciNetCrossRef
5.
go back to reference Yuille AL, Rangarajan A (2002) The concave-convex procedure (cccp). In: Advances in neural information processing systems. MIT Press, pp 1033–1040 Yuille AL, Rangarajan A (2002) The concave-convex procedure (cccp). In: Advances in neural information processing systems. MIT Press, pp 1033–1040
6.
go back to reference Minka TP (2001) Expectation propagation for approximate bayesian inference. In: Proceedings of the seventeenth conference on uncertainty in artificial intelligence. Morgan Kaufmann Publishers Inc. pp 362–369 Minka TP (2001) Expectation propagation for approximate bayesian inference. In: Proceedings of the seventeenth conference on uncertainty in artificial intelligence. Morgan Kaufmann Publishers Inc. pp 362–369
7.
go back to reference Seeger MW (2008) Bayesian inference and optimal design for the sparse linear model. J Mach Learn Res 9:759–813 JuneMathSciNetMATH Seeger MW (2008) Bayesian inference and optimal design for the sparse linear model. J Mach Learn Res 9:759–813 JuneMathSciNetMATH
8.
go back to reference Schniter P, Rangan S, Fletcher AK (2016) Vector approximate message passing for the generalized linear model. In: 50th asilomar conference on signals. Systems and computers. Pacific Grove, CA, USA, Nov, , pp 1525–1529 Schniter P, Rangan S, Fletcher AK (2016) Vector approximate message passing for the generalized linear model. In: 50th asilomar conference on signals. Systems and computers. Pacific Grove, CA, USA, Nov, , pp 1525–1529
9.
go back to reference Ko Y-J, Seeger MW (2016) Expectation propagation for rectified linear poisson regression. In: Asian conference on machine learning, ser. Proceedings of machine learning research, vol 45, Hong Kong, pp 253–268 Ko Y-J, Seeger MW (2016) Expectation propagation for rectified linear poisson regression. In: Asian conference on machine learning, ser. Proceedings of machine learning research, vol 45, Hong Kong, pp 253–268
10.
go back to reference Altmann Y, Perelli A, Davies ME (2019) Expectation-Propagation algorithms for linear regression with Poisson noise: application to photon-limited spectral unmixing. In: IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Brighton, UK Altmann Y, Perelli A, Davies ME (2019) Expectation-Propagation algorithms for linear regression with Poisson noise: application to photon-limited spectral unmixing. In: IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), Brighton, UK
12.
go back to reference Figueiredo M, Bioucas-Dias J (2010) Restoration of Poissonian images using alternating direction optimization. IEEE Trans Image Proc 19(12):3133–3145MathSciNetCrossRef Figueiredo M, Bioucas-Dias J (2010) Restoration of Poissonian images using alternating direction optimization. IEEE Trans Image Proc 19(12):3133–3145MathSciNetCrossRef
13.
go back to reference Altmann Y, Maccarone A, Halimi A, McCarthy A, Buller GS, McLaughlin S (2016) Efficient range estimation and material quantification from multispectral lidar waveforms. In: Proceeding sensor signal processing for defence (SSPD) conference. Edinburgh, UK Altmann Y, Maccarone A, Halimi A, McCarthy A, Buller GS, McLaughlin S (2016) Efficient range estimation and material quantification from multispectral lidar waveforms. In: Proceeding sensor signal processing for defence (SSPD) conference. Edinburgh, UK
14.
go back to reference Tobin R, Altmann Y, Ren X, McCarthy A, Lamb RA, McLaughlin S, Buller GS (2017) Comparative study of sampling strategies for sparse photon multispectral lidar imaging: towards mosaic filter arrays. J Opt 19(9):094006CrossRef Tobin R, Altmann Y, Ren X, McCarthy A, Lamb RA, McLaughlin S, Buller GS (2017) Comparative study of sampling strategies for sparse photon multispectral lidar imaging: towards mosaic filter arrays. J Opt 19(9):094006CrossRef
15.
go back to reference Févotte C, Dobigeon N (2015) Nonlinear hyperspectral unmixing with robust nonnegative matrix factorization. IEEE Trans Image Proc 24(12):4810–4819MathSciNetCrossRef Févotte C, Dobigeon N (2015) Nonlinear hyperspectral unmixing with robust nonnegative matrix factorization. IEEE Trans Image Proc 24(12):4810–4819MathSciNetCrossRef
16.
go back to reference Brooks S (2011) Handbook of markov chain monte carlo, Chapman & Hall/CRC Handbooks of Modern Statistical Methods. Taylor & Francis Brooks S (2011) Handbook of markov chain monte carlo, Chapman & Hall/CRC Handbooks of Modern Statistical Methods. Taylor & Francis
17.
go back to reference Bouchard-Côté A, Vollmer SJ, Doucet A (2018) The bouncy particle sampler: a nonreversible rejection-free markov chain monte carlo method. J Am Stat Assoc 113(522):855–867MathSciNetCrossRef Bouchard-Côté A, Vollmer SJ, Doucet A (2018) The bouncy particle sampler: a nonreversible rejection-free markov chain monte carlo method. J Am Stat Assoc 113(522):855–867MathSciNetCrossRef
Metadata
Title
Robust Linear Regression and Anomaly Detection in the Presence of Poisson Noise Using Expectation-Propagation
Authors
Yoann Altmann
Dan Yao
Stephen McLaughlin
Mike E. Davies
Copyright Year
2021
Publisher
Springer Singapore
DOI
https://doi.org/10.1007/978-981-15-9199-0_14

Premium Partners