Abstract
We present a novel method for online inference of real-valued quantities on a large network from very sparse measurements. The target application is a large scale system, like e.g. a traffic network, where a small varying subset of the variables is observed, and predictions about the other variables have to be continuously updated. A key feature of our approach is the modeling of dependencies between the original variables through a latent binary Markov random field. This greatly simplifies both the model selection and its subsequent use. We introduce the mirror belief propagation algorithm, that performs fast inference in such a setting. The offline model estimation relies only on pairwise historical data and its complexity is linear w.r.t. the dataset size. Our method makes no assumptions about the joint and marginal distributions of the variables but is primarily designed with multimodal joint distributions in mind. Numerical experiments demonstrate both the applicability and scalability of the method in practice.
Similar content being viewed by others
References
Baxter, R.: Exactly solved models in statistical mechanics. Dover Publications (2008)
Beyer, K., Goldstein, J., Ramakrishnan, R., Shaft, U.: When is “nearest neighbor” meaningful? In Database Theory-ICDT’99, pp 217–235. Springer (1999)
Bickson, D.: Gaussian Belief Propagation: Theory and Application. PhD thesis, Hebrew University of Jerusalem (2008)
Bilmes, J.: On soft evidence in Bayesian networks. Technical report, University of Washington (2004)
Boyen, X.: Inference and Learning in Complex Stochastic Processes. PhD thesis, Stanford University, Computer Science Department, 229 (2002)
Chan, H., Darwiche, A.: On the revision of probabilistic beliefs using uncertain evidence. Artif. Intell. 163(1), 67–90 (2005)
Cocco, S., Monasson, R.: Adaptive cluster expansion for the inverse Ising problem: convergence, algorithm and tests. J. Stat. Phys. 147(2), 252–314 (2012)
Cover, T., Hart, P.: Nearest neighbor pattern classification. IEEE Trans. Inf. Theory 13(1), 21–27 (1967)
Darroch, J., Ratcliff, D.: Generalized iterative scaling for log-linear models. The Annals Math. Stat. 43(5), 1470–1480 (1972)
Dempster, A., Laird, N., Rubin, D.: Maximum likelihood from incomplete data via the EM algorithm. J. Royal Stat. Soc. Ser. B (Methodological), 1–38 (1977)
Doucet, A., de Freitas, N., Gordon, N.: An introduction to sequential monte carlo methods. Springer-Verlag, New York (2001)
Friedman, J., Hastie, T., Tibshirani, R.: Sparse inverse covariance estimation with the graphical lasso. Biostatistics 9(3), 432–441 (2008)
Furtlehner, C.: Approximate inverse Ising models close to a Bethe reference point. J. Stat. Mech.: Theory Exp. 2013(09), P09020 (2013)
Furtlehner, C., Han, Y., Lasgouttes, J.-M., Martin, V., Marchal, F., Moutarde, F.: Spatial and temporal analysis of traffic states on large scale networks. In: Procceding of the 13th International IEEE Conference on Intelligent Transportation Systems, 1215–1220 (2010)
Furtlehner, C., Lasgouttes, J.-M., Auger, A.: Learning multiple belief propagation fixed points for real time inference. Phys. A: Stat. Mech. Appl. 389(1), 149–163 (2010)
Han, T.X., Ning, H., Huang, T.S.: Efficient nonparametric belief propagation with application to articulated body tracking. In: Computer Vision and Pattern Recognition, 2006 IEEE Computer Society Conference on, volume 1, 214–221 (2006)
Herrera, J., Work, D., Herring, R., Ban, X., Jacobson, Q., Bayen, A.: Evaluation of traffic data obtained via GPS-enabled mobile phones: The mobile century field experiment. Transp. Res. Part C: Emerg. Technol. 18(4), 568–583 (2010)
Ihler, A., Fischer, J.I., Willsky, A.: Loopy belief propagation: convergence and effects of message errors. J. Mach. Learn. Res. 6, 905–936 (2005)
Jalali, A., Johnson, C., Ravikumar, P.: On learning discrete graphical models using greedy methods. arXiv:1107.3258 (2011)
Jaynes, E.T.: Prior probabilities. IEEE Trans. Syst. Sci. Cybern. 4(3), 227–241 (1968)
Jaynes, E.T.: Probability Theory: The Logic of Science (Vol 1). Cambridge University Press, 2003. ISBN 0521592712
Kschischang, F.R., Frey, B.J., Loeliger, H.: Factor graphs and the sum-product algorithm. IEEE Trans. Inf. Theory 47(2), 498–519 (2001)
Mackay, D.J., Yedidia, J.S., Freeman, W.T., Weiss, Y., et al.: A conversation about the Bethe free energy and sum-product. Available at, http://www.merl.com/publications/TR2001-018/ (2001)
Martin, V.: Modélisation Probabiliste et inférence par l’algorithme belief propagation. PhD thesis, Mines-ParisTech (2013)
Martin, V., Furtlehner, C., Han, Y., Lasgouttes, J.-M.: GMRF Estimation under Spectral and Topological constraints. In: Machine Learning and Knowledge Discovery in Databases, volume 8725 of Lecture Notes in Computer Science, pages 370–385. Springer Berlin Heidelberg (2014)
Mézard, M., Mora, T.: Constraint satisfaction problems and neural networks: A statistical physics perspective. J. Physiology-Paris 103(1-2), 107–113 (2009)
Mézard, M., Parisi, G., Virasoro, M.: Spin glass theory and beyond. World scientific, Singapore (1987)
Min, W., Wynter, L.: Real-time road traffic prediction with spatio-temporal correlations. Transp. Res. Part C 19, 606–616 (2011)
Mooij, J.M., Kappen, H.J.: Sufficient conditions for convergence of the sum-product algorithm. IEEE Trans. Inf. Theory 53(12), 4422–4437 (2007)
Pearl, J.: Probabilistic Reasoning in Intelligent Systems: Network of Plausible Inference. Morgan Kaufmann (1988)
PUMAS project: http://team.inria.fr/pumas/ (in French)
Ravikumar, P., Wainwright, M.J., Lafferty, J.D.: High-dimensional Ising model selection using L 1 regularized logistic regression. Ann. Stat. 38(3), 1287–1319 (2010)
Smith, B.L., Williams, B.M., Keith Oswald, R.: Comparison of parametric and nonparametric models for traffic flow forecasting. Trans. Res, Part C: Emerg. Technol. 10(4), 303–321 (2002)
Sudderth, E., Ihler, A., Isard, M., Freeman, W., Willsky, A.: Nonparametric Belief Propagation. Commun. ACM 53(10), 95–103 (Oct. 2010)
Tatikonda, S., Jordan, M.: Loopy Belief Propagation and Gibbs measures. In: Proceedings of the 18th Conference in Uncertainty in Artificial Intelligence, 493–50 (2002)
Teh, Y.W., Welling, M.: Passing and bouncing messages for generalized inference. Technical report, UCL (2001)
Wainwright, M.J.: Estimating the “wrong” graphical model: benefits in the computation-limited setting. J. Mach. Learn. Res. 7, 1829–1859 (2006)
Welling, M., Teh, Y.W.: Approximate inference in Boltzmann machines. Artif. Intell. 143(1), 19–50 (2003)
Yasuda, M., Tanaka, K.: Approximate learning algorithm in Boltzmann machines. Neural Comput. 21(11), 3130–3178 (2009)
Yedidia, J.S., Freeman, W.T., Weiss, Y.: Constructing free-energy approximations and generalized Belief Propagation algorithms. IEEE Trans. Inf. Theory 51(7), 2282–2312 (2005)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Martin, V., Lasgouttes, JM. & Furtlehner, C. Latent binary MRF for online reconstruction of large scale systems. Ann Math Artif Intell 77, 123–154 (2016). https://doi.org/10.1007/s10472-015-9470-x
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10472-015-9470-x