Abstract
We present a method for constructing a generative model for sets of graphs by adopting a minimum description length approach. The method is posed in terms of learning a generative supergraph model from which the new samples can be obtained by an appropriate sampling mechanism. We commence by constructing a probability distribution for the occurrence of nodes and edges over the supergraph. We encode the complexity of the supergraph using the von-Neumann entropy. A variant of EM algorithm is developed to minimize the description length criterion in which the node correspondences between the sample graphs and the supergraph are treated as missing data.The maximization step involves updating both the node correspondence information and the structure of supergraph using graduated assignment. In the experimental part, we demonstrate the practical utility of our proposed algorithm and show that our generative model gives good graph classification results. Besides, we show how to perform graph clustering with Jensen-Shannon kernel and generate new sample graphs.
This is a preview of subscription content, log in via an institution to check access.
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
Friedman, N., Koller, D.: Being Bayesian about network structure.A Bayesian approach to structure discovery in Bayesian networks. Machine Learning, 95–125 (2003)
Christmas, W.J., Kittler, J., Petrou, M.: Probabilistic feature labeling schemes: modeling compatibility coefficient distribution. Image and Vision Computing 14, 617–625 (1996)
Bagdanov, A.D., Worring, M.: First order Gaussian graphs for efficient structure classification. Pattern Recognition 36, 1311–1324 (2003)
Cootes, T.F., Edwards, G.J., Taylor, C.J.: Active Appearance Models. IEEE PAMI 23, 681–685 (2001)
Luo, B., Hancock, E.R.: A spectral approach to learning structural variations in graphs. Pattern Recognition 39, 1188–1198 (2006)
Torsello, A., Hancock, E.R.: Learning shape-classes using a mixture of tree-unions. IEEE PAMI 28, 954–967 (2006)
Luo, B., Hancock, E.R.: Structural graph matching using the EM alogrithm and singular value decomposition. IEEE PAMI 23, 1120–1136 (2001)
White, D., Wilson, R.C.: Parts based generative models for graphs. In: ICPR, pp. 1–4 (2008)
Rissanen, J.: Modelling by Shortest Data Description. Automatica, 465–471 (1978)
Rissanen, J.: Stochastic complexity in statistical inquiry. World Scientific, Singapore (1989)
Passerini, F., Severini, S.: The von neumann entropy of networks, arXiv:0812.2597 (2008)
Cover, T., Thomas, J.: Elements of Information Theory. John Wiley&Sons, New York (1991)
Grunwald, P.: Minimum Description Length Tutorial. In: Advances in Minimum Description Length: Theory and Applications (2005)
Bridle, J.S.: Training stochastic model recognition algorithms as networks can lead to maximum mutual information estimation of parameters. In: Advances in Neural Information Processing Systems, vol. 2, pp. 211–217 (1990)
Gold, S., Rangarajan, A.: A Graduated Assignment Algorithm for Graph Matching. IEEE PAMI 18, 377–388 (1996)
Figueiredo, M.A.T., Jain, A.K.: Unsupervised learning of finite mixture models. IEEE PAMI 24, 381–396 (2002)
Wilson, R.C., Zhu, P.: A study of graph spectra for comparing graphs and trees. Pattern Recognition 41, 2833–2841 (2008)
Shi, J., Malik, J.: Normalized cuts and image segmentation. In: CVPR, pp. 731–737 (1997)
Robles-Kelly, A., Hancock, E.R.: A riemannian approach to graph embedding. Pattern Recognition 40, 1042–1056 (2007)
Nene, S.A., Nayar, S.K., Murase, H.: Columbiaobjectimagelibrary (COIL100). Technical Report CUCS-006-96. Department of Computer Science, Columbia University (1996)
Lowe, D.G.: Distinctive image features from scale invariant keypoints. IJCV 99, 91–110 (2004)
Wilson, R.C., Hancock, E.R.: Structural matching by discrete relaxation. IEEE PAMI 19, 634–648 (1997)
Han, L., Wilson, R.C., Hancock, E.R.: A Supergraph-based Generative Model. In: ICPR, pp. 1566–1569 (2010)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2011 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Han, L., Hancock, E.R., Wilson, R.C. (2011). An Information Theoretic Approach to Learning Generative Graph Prototypes. In: Pelillo, M., Hancock, E.R. (eds) Similarity-Based Pattern Recognition. SIMBAD 2011. Lecture Notes in Computer Science, vol 7005. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-24471-1_10
Download citation
DOI: https://doi.org/10.1007/978-3-642-24471-1_10
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-24470-4
Online ISBN: 978-3-642-24471-1
eBook Packages: Computer ScienceComputer Science (R0)