Abstract
In many real-world problems, the interest is in learning about unknowns from observed data. One widely adopted approach to learning is to employ Bayesian inference. Over the last half-century, Monte Carlo-based methods have kept the main stage of Bayesian reasoning, and they will not relinquish it any time soon. With these methods, Bayesianism has enjoyed tremendous success as they made possible the computation of posterior distributions of unknowns from very challenging models. In the new age of signal processing, the main problems of concern are those with large numbers of unknowns (complex systems) and/or large amounts of data (big data). This raises the concern that Bayesian inference may become computationally incapable of handling them because of the sheer size and complexity of the studied systems. This chapter discusses the main challenges related to the addressed problem and recent advances in the theory and practice of a class of Monte Carlo methods, adaptive importance sampling (AIS), for dealing exactly with the types of problems where the numbers of unknowns and/or data are large. The focus of the chapter will be on new schemes for adaptive importance sampling with emphasis on strategies for adaptive learning and for stable weight computation, model selection, and application of the theory to case-studies including reconstruction of gene regulatory (life science problem) and inference of demographic rates of penguin populations (ecological problem).
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
- 2.
Asymptotic methods, multiple quadrature approaches, and subregion adaptive integration (Evans and Swartz 1995) are also used for approximation of integrals but cannot be applied in high-dimensional setups and only MC approaches become feasible in many practical applications.
- 3.
Note that here we discuss initialization strategies for Bayesian models with only unknown static parameters for simplicity in the notation and presentation.
References
Adali T, Haykin S (2010) Adaptive signal processing: next generation solutions, vol 55. Wiley
Andrieu C, Davy M, Doucet A (2001) Improved auxiliary particle filtering: Applications to time-varying spectral analysis. In: Proceedings of the 11th IEEE signal processing workshop on statistical signal processing (Cat. No. 01TH8563). IEEE, pp 309–312
Andrieu C, Doucet A, Holenstein R (2010) Particle Markov chain Monte Carlo methods. J Roy Stat Soc B 72(3):269–342
Barabasi AL, Oltvai ZN (2004) Network biology: Understanding the cell’s functional organization. Nat Rev Genet 5:101–113
Bengtsson T, Snyder C, Nychka D (2003) Toward a nonlinear ensemble filter for high dimensional systems. J Geophys Res 108
Bernardo JM, Smith AF (2001) Bayesian theory. Wiley
Berzuini C, Best NG, Gilks WR, Larizza C (1997) Dynamic conditional independence models and Markov Chain Monte Carlo methods. J Am Stat Assoc 92:1403–1412
Binford TO, Levitt TS, Mann WB (2013) Bayesian inference in model-based machine vision. Preprint.‘arXiv:13042720
Bojesen TA (2018) Policy-guided Monte Carlo: Reinforcement-learning Markov chain dynamics. Phys Rev E 98(6):063303
Boyd S, Vandenberghe L (2004) Convex optimization. Cambridge University Press
Brooks S, Gelman A, Jones GL, Meng XL (eds) (2011) Handbook of Markov Chain Monte Carlo. CRC Press
Bugallo MF, Elvira V, Martino L, Luengo D, Míguez J, Djurić PM (2017) Adaptive importance sampling: The past, the present, and the future. Signal Process Mag 34:60–79
Bugallo MF, Taşdemir Ç, Djurić PM (2015) Estimation of gene expression by a bank of particle filters. In: 2015 23rd European signal processing conference (EUSIPCO). IEEE, pp 494–498
Bui T, Hernández-Lobato D, Hernandez-Lobato J, Li Y, Turner R (2016) Deep gaussian processes for regression using approximate expectation propagation. In: International conference on machine learning, pp 1472–1481
Cappé O, Douc R, Guillin A, Marin J, Robert CP (2008a) Adaptive importance sampling in general mixture classes. Stat Comput 18(4):447–459
Cappé O, Douc R, Guillin A, Marin JM, Robert CP (2008b) Adaptive importance sampling in general mixture classes. Stat Comput 18:447–459
Cappé O, Guillin A, Marin J, Robert CP (2004) Population Monte Carlo. J Comput Graph Stat 13(4):907–929
Casella G, Berger RL (2002) Statistical inference. Duxbury, Pacific Grove, CA (USA)
Chang C, Ding Z, Hung YS, Fung PCW (2008) Fast network component analysis (fastnca) for gene regulatory network reconstruction from microarray data. Bioinformatics 24(11):1349–1358
Chen H, Tiňo P, Yao X (2009) Predictive ensemble pruning by expectation propagation. IEEE Trans Knowl Data Eng (7):999–1013
Claeskens G, Hjort NL (2008) Model selection and model averaging. Cambridge University Press
Cornuet J, Marin J, Mira A, Robert CP (2012) Adaptive multiple importance sampling. Scand J Stat 39(4):798–812
Ding J, Tarokh V, Yang Y (2018) Model selection techniques: An overview. IEEE Signal Process Mag 35(6):16–34
Djurić PM, Lu T, Bugallo MF (2007) Multiple particle filtering. In: 2007 IEEE international conference on acoustics, speech and signal processing-ICASSP’07, vol 3. IEEE, pp III–1181
Doucet A, Johansen AM (2009) A tutorial on particle filtering and smoothing: Fifteen years later. Handbook of nonlinear filtering, vol 12(656-704), p 3
El-Laham Y, Djurić PM, Bugallo MF (2019) A variational adaptive population importance sampler. In: ICASSP 2019-2019 IEEE international conference on acoustics, speech and signal processing (ICASSP). IEEE, pp 5052–5056
El-Laham Y, Elvira V, Bugallo MF (2018) Robust covariance adaptation in adaptive importance sampling. IEEE Signal Process Lett 25
El-Laham Y, Yang L, Lynch HJ, Djurić PM, Bugallo MF (2021) Particle Gibbs sampling for regime-switching state-space models. In: IEEE international conference on acoustics, speech and signal processing (ICASSP)
Evans M, Swartz T (1995) Methods for approximating integrals in statistics with special emphasis on Bayesian integration problems. Stat Sci, 254–272
Fearnhead P (2008) Computational methods for complex stochastic systems: A review of some alternatives to MCMC. Stat Comput 18(2):151–171
Gilks WR, Berzuini C (2001) Following a moving target - Monte Carlo inference for dynamic Bayesian model. J Roy Stat Soc B 63:127–146
Godsill SJ, Rayner PJW (1998) Digital audio restoration, 1st edn. Springer, London (UK)
Gordon NJ, Salmond DJ, Smith AFM (1993) Novel approach to nonlinear/non-Gaussian Bayesian state estimation. In: IEE proceedings F (radar and signal processing), vol 140, pp 107–113
Graves A (2016) Stochastic backpropagation through mixture density distributions. Preprint. arXiv:160705690
Green PJ (1995) Reversible jump Markov chain Monte Carlo computation and Bayesian model determination. Biometrika 4:711–732
Haykin SS (1994) Blind deconvolution. Prentice Hall, Englewood Cliffs, NJ (USA)
Heng J, Bishop AN, Deligiannidis G, Doucet A (2017) Controlled sequential Monte Carlo. Preprint. arXiv:170808396
Hinne M, Heskes T, Beckmann CF, van Gerven MAJ (2013) Bayesian inference of structural brain networks. NeuroImage 66:543–552
Hong M, Bugallo MF, Djurić PM (2010) Joint model selection and parameter estimation by population Monte Carlo simulation. IEEE J Sel Top Signal Process 4:526–539
Hutt R (2016) What are the 10 biggest global challenges? In: World economic forum annual meeting. https://www.weforum.org/agenda/2016/01/what-are-the-10-biggest-global-challenges/
Iloska M, El-Laham Y, Bugallo MF (2020) A particle Gibbs sampling approach to topology inference in gene regulatory networks. In: IEEE international conference on acoustics, speech and signal processing (ICASSP)
Jenkins CR, Peacock JA (2011) The power of Bayesian evidence in astronomy. Mon Not Roy Astron Soc 413:2895–2905
Jordan M, Kleinberg J (2006) Information science and statistics, vol 4
Karlsson R (2005) Particle filtering for positioning and tracking applications. PhD thesis, Linköping University Electronic Press
Kay SM (1993) Fundamentals of statistical signal processing: estimation theory. Prentice Hall, Upper Saddle River, NJ (USA)
Kilbinger M, Wraith D, Robert CP, Benabed K, Cappé O, Cardoso JF, Fort G, Prunet S, Bouchet FR (2010) Bayesian model comparison in cosmology with population Monte Carlo. Mon Not Roy Astron Soc 405:2381–2390
Li T, Bolic M, Djuric PM (2015) Resampling methods for particle filtering: classification, implementation, and strategies. IEEE Signal Process Mag 32(3):70–86
Liu J, West M (2001) Combined parameter and state estimation in simulation-based filtering. In: Sequential Monte Carlo methods in practice. Springer, pp 197–223
Lopes HF, Tsay RS (2011) Particle filters and Bayesian inference in financial econometrics. J Forecast 30:168–209
Marin JM, Robert CP (2010) A practical approach to computational Bayesian statistics. Springer
Martino L, Elvira V, Luengo D, Corander J (2015) An adaptive population importance sampler: Learning from uncertainty. IEEE Trans Signal Process 63(16):4422–4437
Martino L, Elvira V, Luengo D, Corander J (2017) Layered adaptive importance sampling. Stat Comput 27(3):599–623
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
Minsker S, Srivastava S, Lin L, Dunson D (2014) Scalable and robust Bayesian inference via the median posterior. In: International conference on machine learning, pp 1656–1664
Morita S, Thall PF, Müller P (2008) Determining the effective sample size of a parametric prior. Biometrics 64(2):595–602
Murphy KP (2012) Machine learning: A probabilistic perspective. The MIT Press
Naesseth CA, Linderman SW, Ranganath R, Blei DM (2017) Variational sequential Monte Carlo. Preprint. arXiv:170511140
Neiswanger C Wand Wang, Xing E (2013) Asymptotically exact, embarrassingly parallel MCMC. Preprint. arXiv:13114780
Orchard ME, Vachtsevanos GJ (2007) A particle filtering-based framework for real-time fault diagnosis and failure prognosis in a turbine engine. In: 2007 mediterranean conference on control & automation. IEEE, pp 1–6
Ortega A, Frossard P, Kovačević J, Moura JMF, Vandergheynst P (2018) Graph signal processing: Overview, challenges, and applications. Proc IEEE 106(5):808–828
Plummer M (2003) JAGS: A program for analysis of Bayesian graphical models using Gibbs sampling. In: Proceedings of the 3rd international workshop on distributed statistical computing, Vienna, Austria, vol 124, p 10
Portier F, Delyon B (2018) Asymptotic optimality of adaptive importance sampling. In: Advances in neural information processing systems, pp 3134–3144
Putman RJ (1995) Ethical considerations and animal welfare in ecological field studies. Biodivers Conserv 4(8):903–915
Rabin J, Peyré G, Delon J, Bernot M (2011) Wasserstein barycenter and its application to texture mixing. In: International conference on scale space and variational methods in computer vision. Springer, pp 435–446
Ryu EK (2016) Convex optimization for monte carlo: Stochastic optimization for importance sampling. PhD thesis, Stanford University
Ryu EK, Boyd SP (2014) Adaptive importance sampling via stochastic convex programming. Preprint. arXiv:14124845
Scharf LL (1991) Statistical signal processing. Addison-Wesley, Reading, MA (USA)
Srivastava S, Cevher V, Dinh Q, Dunson D (2015) WASP: Scalable Bayes via barycenters of subset posteriors. In: Artificial intelligence and statistics, pp 912–920
Taşdemir Ç, Bugallo MF, Djurić PM (2017) A particle-based approach for topology estimation of gene networks. In: 2017 IEEE 7th International workshop on computational advances in multi-sensor adaptive processing (CAMSAP). IEEE, pp 1–5
Van Trees HL (2004) Detection, estimation, and modulation theory, part I: detection, estimation, and linear modulation theory. Wiley
von Toussaint U (2011) Bayesian inference in physics. Rev Mod Phys 83:943–999
Wang X, Dunson DB (2013) Parallel MCMC via Weierstrass sampler, vol 24. Preprint. arXiv:13124605
Wang X, Li T, Sun S, Corchado J (2017) A survey of recent advances in particle filters and remaining challenges for multitarget tracking. Sensors 17(12):2707
Williams BK, Nichols JD, Conroy MJ (2002) Analysis and management of animal populations. Academic Press
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this chapter
Cite this chapter
Bugallo, M.F. (2023). Recent Advances in Bayesian Inference for Complex Systems. In: Greco, M.S., Cassioli, D., Ullo, S.L., Lyons, M.J. (eds) Women in Telecommunications. Women in Engineering and Science. Springer, Cham. https://doi.org/10.1007/978-3-031-21975-7_4
Download citation
DOI: https://doi.org/10.1007/978-3-031-21975-7_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-21974-0
Online ISBN: 978-3-031-21975-7
eBook Packages: EngineeringEngineering (R0)