Abstract
Markov networks are models for compactly representing complex probability distributions. They are composed by a structure and a set of numerical weights. The structure qualitatively describes independences in the distribution, which can be exploited to factorize the distribution into a set of compact functions. A key application for learning structures from data is to automatically discover knowledge. In practice, structure learning algorithms focused on “knowledge discovery” present a limitation: they use a coarse-grained representation of the structure. As a result, this representation cannot describe context-specific independences. Very recently, an algorithm called CSPC was designed to overcome this limitation, but it has a high computational complexity. This work tries to mitigate this downside presenting CSGS, an algorithm that uses the Grow-Shrink strategy for reducing unnecessary computations. On an empirical evaluation, the structures learned by CSGS achieve competitive accuracies and lower computational complexity with respect to those obtained by CSPC.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Bromberg, F., Margaritis, D., Honavar, V.: Efficient Markov network structure discovery using independence tests. Journal of Artificial Intelligence Research 35(2), 449 (2009)
Edera, A., Schlüter, F., Bromberg, F.: Learning markov networks with context-specific independences. In: The 25th International Conference on Tools with Artificial Intelligence, Herndon, VA, USA, November 4-6, pp. 553–560. IEEE (2013)
Edera, A., Schlüter, F., Bromberg, F.: Learning Markov networks structures constrained by context-specific independences. viXra submission 1405.0222v1 (2014), http://viXra.org/abs/1405.0222
Haaren, J.V., Davis, J.: Markov network structure learning: A randomized feature generation approach. In: Proceedings of the Twenty-Sixth National Conference on Artificial Intelligence. AAAI Press (2012)
Hammersley, J.M., Clifford, P.: Markov fields on finite graphs and lattices (1971) (unpublished manuscript)
Højsgaard, S.: Yggdrasil: a statistical package for learning split models. In: Proceedings of the Sixteenth Conference on Uncertainty in Artificial Intelligence, pp. 274–281. Morgan Kaufmann Publishers Inc. (2000)
Højsgaard, S.: Statistical inference in context specific interaction models for contingency tables. Scandinavian Journal of Statistics 31(1), 143–158 (2004)
Koller, D., Friedman, N.: Probabilistic Graphical Models: Principles and Techniques. MIT Press, Cambridge (2009)
Lauritzen, S.L.: Graphical models. Oxford University Press (1996)
Lowd, D., Davis, J.: Learning Markov network structure with decision trees. In: 2010 IEEE 10th International Conference on Data Mining (ICDM), pp. 334–343. IEEE (2010)
Lowd, D., Davis, J.: Improving markov network structure learning using decision trees. Journal of Machine Learning Research 15, 501–532 (2014), http://jmlr.org/papers/v15/lowd14a.html
Margaritis, D., Thrun, S.: Bayesian network induction via local neighborhoods. Tech. rep., DTIC Document (2000)
Moore, A., Lee, M.S.: Cached Suficient Statistics for Efficient Machine Learning with Large Datasets. Journal of Artificial Intelligence Research 8, 67–91 (1998)
Pearl, J.: Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference, 1st edn. Morgan Kaufmann Publishers Inc. (1988)
Schlüter, F., Bromberg, F., Edera, A.: The IBMAP approach for Markov network structure learning. Annals of Mathematics and Artificial Intelligence, 1–27 (2014), http://dx.doi.org/10.1007/s10472-014-9419-5
Smith, V.A., Yu, J., Smulders, T.V., Hartemink, A.J., Jarvis, E.D.: Computational inference of neural information flow networks. PLoS Computational Biology 2(11), e161 (2006)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this paper
Cite this paper
Edera, A., Strappa, Y., Bromberg, F. (2014). The Grow-Shrink Strategy for Learning Markov Network Structures Constrained by Context-Specific Independences. In: Bazzan, A., Pichara, K. (eds) Advances in Artificial Intelligence -- IBERAMIA 2014. IBERAMIA 2014. Lecture Notes in Computer Science(), vol 8864. Springer, Cham. https://doi.org/10.1007/978-3-319-12027-0_23
Download citation
DOI: https://doi.org/10.1007/978-3-319-12027-0_23
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-12026-3
Online ISBN: 978-3-319-12027-0
eBook Packages: Computer ScienceComputer Science (R0)