Bayesian Analysis

Marginal Pseudo-Likelihood Learning of Discrete Markov Network Structures

Johan Pensar, Henrik Nyman, Juha Niiranen, and Jukka Corander

Full-text: Open access

Enhanced PDF (265 KB)

Abstract

Markov networks are a popular tool for modeling multivariate distributions over a set of discrete variables. The core of the Markov network representation is an undirected graph which elegantly captures the dependence structure over the variables. Traditionally, the Bayesian approach of learning the graph structure from data has been done under the assumption of chordality since non-chordal graphs are difficult to evaluate for likelihood-based scores. Recently, there has been a surge of interest towards the use of regularized pseudo-likelihood methods as such approaches can avoid the assumption of chordality. Many of the currently available methods necessitate the use of a tuning parameter to adapt the level of regularization for a particular dataset. Here we introduce the marginal pseudo-likelihood which has a built-in regularization through marginalization over the graph-specific nuisance parameters. We prove consistency of the resulting graph estimator via comparison with the pseudo-Bayesian information criterion. To identify high-scoring graph structures in a high-dimensional setting we design a two-step algorithm that exploits the decomposable structure of the score. Using synthetic and existing benchmark networks, the marginal pseudo-likelihood method is shown to perform favorably against recent popular structure learning methods.

Article information

Source
Bayesian Anal., Volume 12, Number 4 (2017), 1195-1215.

Dates
First available in Project Euclid: 31 October 2016

Permanent link to this document
https://projecteuclid.org/euclid.ba/1477918728

Digital Object Identifier
doi:10.1214/16-BA1032

Mathematical Reviews number (MathSciNet)
MR3724983

Zentralblatt MATH identifier
1384.62178

Keywords
Markov networks structure learning pseudo-likelihood non-chordal graph Bayesian inference regularization

Rights
Creative Commons Attribution 4.0 International License.

Citation

Pensar, Johan; Nyman, Henrik; Niiranen, Juha; Corander, Jukka. Marginal Pseudo-Likelihood Learning of Discrete Markov Network Structures. Bayesian Anal. 12 (2017), no. 4, 1195--1215. doi:10.1214/16-BA1032. https://projecteuclid.org/euclid.ba/1477918728


Export citation

References

  • Abellán, J., Gómez-Olmedo, M., and Moral, S. (2006). “Some variations on the PC algorithm.” In Proceedings of the 3rd European Workshop on Probabilistic Graphical Models, 1–8.
  • Akaike, H. (1974). “A new look at the statistical model identification.” IEEE Transactions on Automatic Control, 19: 716–723.
    Zentralblatt MATH: 0314.62039
    Digital Object Identifier: doi:10.1109/TAC.1974.1100705
  • Anandkumar, A., Tan, V. Y. F., Huang, F., and Willsky, A. S. (2012). “High-dimensional structure estimation in Ising models: Local separation criterion.” The Annals of Statistics, 40: 1346–1375.
    Mathematical Reviews (MathSciNet): MR3015028
    Zentralblatt MATH: 1297.62124
    Digital Object Identifier: doi:10.1214/12-AOS1009
    Project Euclid: euclid.aos/1344610586
  • Aurell, E. and Ekeberg, M. (2012). “Inverse Ising inference using all the data.” Physical Review Letters, 108: 090201.
  • Barber, R. F. and Drton, M. (2015). “High-dimensional Ising model selection with Bayesian information criteria.” Electronic Journal of Statistics, 9(1): 567–607.
    Zentralblatt MATH: 1309.62050
    Digital Object Identifier: doi:10.1214/15-EJS1012
  • Bartlett, M. and Cussens, J. (2013). “Advances in Bayesian Network Learning using Integer Programming.” In Proceedings of the 29th Conference on Uncertainty in Artificial Intelligence, 182–191.
  • Berg, J., Järvisalo, M., and Malone, B. (2014). “Learning optimal bounded treewidth Bayesian networks via maximum satisfiability.” In Proceedings of the 17th Conference on Artificial Intelligence and Statistics, 86–95.
  • Besag, J. (1975). “Statistical analysis of non-lattice data.” Journal of the Royal Statistical Society. Series D (The Statistician), 24: 179–195.
  • Bromberg, F., Margaritis, D., and Honavar, V. (2009). “Efficient Markov network structure discovery using independence tests.” Journal of Artificial Intelligence Research, 35: 449–485.
    Zentralblatt MATH: 1192.68817
  • Chow, C. and Liu, C. (1968). “Approximating discrete probability distributions with dependence trees.” IEEE Transactions on Information Theory, 14(3): 462–467.
    Zentralblatt MATH: 0165.22305
    Digital Object Identifier: doi:10.1109/TIT.1968.1054142
  • Corander, J., Janhunen, T., Rintanen, J., Nyman, H., and Pensar, J. (2013). “Learning chordal Markov networks by constraint satisfaction.” In Advances in Neural Information Processing Systems 26, 1349–1357.
  • Csiszár, I. and Talata, Z. (2006). “Consistent estimation of the basic neighborhood of Markov random fields.” Annals of Statistics, 34: 123–145.
    Zentralblatt MATH: 1102.62105
    Digital Object Identifier: doi:10.1214/009053605000000912
    Project Euclid: euclid.aos/1146576258
  • Ekeberg, M., Lövkvist, C., Lan, Y., Weigt, M., and Aurell, E. (2013). “Improved contact prediction in proteins: Using pseudolikelihoods to infer Potts models.” Physical Review E, 87: 012707.
  • Heckerman, D., Chickering, D. M., Meek, C., Rounthwaite, R., and Kadie, C. (2000). “Dependency networks for inference, collaborative filtering, and data visualization.” Journal of Machine Learning Research, 1: 49–75.
    Zentralblatt MATH: 1008.68132
  • Heckerman, D., Geiger, D., and Chickering, D. M. (1995). “Learning Bayesian networks: The combination of knowledge and statistical data.” Machine Learning, 20: 197–243.
    Zentralblatt MATH: 0831.68096
  • Höfling, H. and Tibshirani, R. (2009). “Estimation of sparse binary pairwise Markov networks using pseudo-likelihoods.” Journal of Machine Learning Research, 10: 883–906.
    Zentralblatt MATH: 1245.62121
  • Ji, C. and Seymour, L. (1996). “A consistent model selection procedure for Markov random fields based on penalized pseudolikelihood.” Annals of Applied Probability, 6: 423–443.
    Zentralblatt MATH: 0856.62082
    Digital Object Identifier: doi:10.1214/aoap/1034968138
    Project Euclid: euclid.aoap/1034968138
  • Koller, D. and Friedman, N. (2009). Probabilistic Graphical Models: Principles and Techniques. MIT Press.
    Zentralblatt MATH: 1183.68483
  • Lauritzen, S. L. (1996). Graphical Models. Oxford: Oxford University Press.
    Zentralblatt MATH: 0907.62001
  • Lee, S.-I., Ganapathi, V., and Koller, D. (2006). “Efficient structure learning of Markov networks using $\ell_{1}$-regularization.” In Advances in Neural Information Processing Systems 19, 817–824.
  • Liu, Q. and Ihler, A. T. (2012). “Distributed parameter estimation via pseudo-likelihood.” In Proceedings of the 29th International Conference on Machine Learning, 1487–1494.
  • Lowd, D. and Davis, J. (2014). “Improving Markov network structure learning using decision trees.” Journal of Machine Learning Research, 15: 501–532.
    Zentralblatt MATH: 1317.68173
  • Meinshausen, N. and Bühlmann, P. (2006). “High-dimensional graphs and variable selection with the lasso.” The Annals of Statistics, 34(3): 1436–1462.
    Zentralblatt MATH: 1113.62082
    Digital Object Identifier: doi:10.1214/009053606000000281
    Project Euclid: euclid.aos/1152540754
  • Mizrahi, Y. D., Denil, M., and de Freitas, N. (2014). “Linear and parallel learning of Markov random fields.” In Proceedings of the 31st International Conference on Machine Learning, 199–207.
  • Murphy, K. P. (2001). “The Bayes net toolbox for MATLAB.” Computing Science and Statistics, 33: 1024–1034.
  • Nyman, H., Pensar, J., Koski, T., and Corander, J. (2014). “Stratified graphical models – context-specific independence in graphical models.” Bayesian Analysis, 9(4): 883–908.
    Zentralblatt MATH: 1327.62030
    Digital Object Identifier: doi:10.1214/14-BA882
  • Parviainen, P., Farahani, H., and Lagergren, J. (2014). “Learning bounded tree-width Bayesian networks using integer linear programming.” In Proceedings of the 17th Conference on Artificial Intelligence and Statistics, 751–759.
  • Pensar, J., Nyman, H., Niiranen, J., and Corander, J. (2016). “Supplementary appendix to “Marginal pseudo-likelihood learning of discrete Markov network structures”.” Bayesian Analysis.
    Zentralblatt MATH: 1384.62178
    Digital Object Identifier: doi:10.1214/16-BA1032
  • Pensar, J., Nyman, H., Koski, T., and Corander, J. (2015). “Labeled directed acyclic graphs: A generalization of context-specific independence in directed graphical models.” Data Mining and Knowledge Discovery, 29(2): 503–533.
  • Pietra, S. D., Pietra, V. D., and Lafferty, J. (1997). “Inducing features of random fields.” IEEE Transactions on Pattern Analysis and Machine Intelligence, 19: 380–393.
  • Ravikumar, P., Wainwright, M. J., and Lafferty, J. D. (2010). “High-dimensional Ising model selection using $\ell_{1}$-regularized logistic regression.” Annals of Statistics, 38: 1287–1319.
    Zentralblatt MATH: 1189.62115
    Digital Object Identifier: doi:10.1214/09-AOS691
    Project Euclid: euclid.aos/1268056617
  • Schmidt, M. (2010). “L1General.” https://www.cs.ubc.ca/~schmidtm/Software/L1General.html.
  • Schwarz, G. (1978). “Estimating the dimension of a model.” Annals of Statistics, 6: 461–464.
    Zentralblatt MATH: 0379.62005
    Digital Object Identifier: doi:10.1214/aos/1176344136
    Project Euclid: euclid.aos/1176344136
  • Scutari, M. (2010). “Learning Bayesian networks with the bnlearn R package.” Journal of Statistical Software, 35(3): 1–22.
  • Spirtes, P., Glymour, C., and Scheines, R. (2000). Causation, Prediction, and Search. MIT Press, 2nd edition.
    Zentralblatt MATH: 0806.62001
  • Tsamardinos, I., Aliferis, C., Statnikov, A., and Statnikov, E. (2003). “Algorithms for large scale Markov blanket discovery.” In The 16th International FLAIRS Conference, 376–380.
  • Tsamardinos, I., Brown, L. E., and Aliferis, C. F. (2006). “The max–min hill-climbing Bayesian network structure learning algorithm.” Machine Learning, 65: 31–78.
  • Whittaker, J. (1990). Graphical Models in Applied Multivariate Statistics. Chichester: Wiley.
    Zentralblatt MATH: 0732.62056

Supplemental materials

  • Appendix: Supplementary Appendix to “Marginal Pseudo-Likelihood Learning of Discrete Markov Network Structures”. The appendix contains a proof of the consistency theorem, pseudocode of the search algorithms, and detailed results from the numerical experiments.
    Digital Object Identifier: doi:10.1214/16-BA1032SUPP
    Supplemental PDF (217 KB)

International Society for Bayesian Analysis

Bayesian Analysis

New content alerts

Email RSS ToC RSS Article
  • You have access to this content.
  • You have partial access to this content.
  • You do not have access to this content.
Turn Off MathJax

What is MathJax?

More like this

+ See more