Large Sample Statistics in the Domain of Graphs

One challenge in bridging the gap between structural and statistical pattern recognition consists in studying combinatorial structures like graphs using probabilistic methods. This contribution presents the structural counterparts of the first and second fundamental theorem in probability, (1) the law of large numbers and (2) the central limit theorem. In addition, we derive characterizations and uniqueness conditions for the mean of graphs. As a special case, we investigate the weighted mean of two graphs. The proposed results establish a sound statistical foundation for unsupervised structural pattern recognition methods.

[1]  M. Fréchet.  Les éléments aléatoires de nature quelconque dans un espace distancié , 1948 .

[2]  Alexander J. Smola,et al.  Learning Graph Matching , 2007, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[3]  Klaus Obermayer,et al.  Consistent Estimator of Median and Mean Graph , 2010, 2010 20th International Conference on Pattern Recognition.

[4]  Klaus Obermayer,et al.  Graph quantization , 2010, Comput. Vis. Image Underst..

[5]  Jiming Peng,et al.  Generalized median graphs and applications , 2009, J. Comb. Optim..

[6]  Shinji Umeyama,et al.  An Eigendecomposition Approach to Weighted Graph Matching Problems , 1988, IEEE Trans. Pattern Anal. Mach. Intell..

[7]  Horst Bunke,et al.  On Median Graphs: Properties, Algorithms, and Applications , 2001, IEEE Trans. Pattern Anal. Mach. Intell..

[8]  Ernest Valveny,et al.  Median graphs: A genetic approach based on new theoretical properties , 2009, Pattern Recognit..

[9]  Jinqiao Duan,et al.  Perspectives in Mathematical Sciences , 2010 .

[10]  Klaus Obermayer,et al.  On the sample mean of graphs , 2008, 2008 IEEE International Joint Conference on Neural Networks (IEEE World Congress on Computational Intelligence).

[11]  Francesc d'ASSIS Serratosa Casanelles,et al.  Theory and algorithms on the median graph , 2009 .

[12]  Fritz Wysotzki,et al.  Central Clustering of Attributed Graphs , 2004, Machine Learning.

[13]  Klaus Obermayer,et al.  Multiple alignment of contact maps , 2009, 2009 International Joint Conference on Neural Networks.

[14]  Horst Bunke,et al.  Self-organizing map for clustering in the graph domain , 2002, Pattern Recognit. Lett..

[15]  Abhishek Bhattacharya,et al.  Statistics on Manifolds with Applications to Shape Spaces , 2009 .

[16]  Abraham Kandel,et al.  Clustering of Web Documents using a Graph Model , 2003, Web Document Analysis.

[17]  Ernest Valveny,et al.  Graph-Based k-Means Clustering: A Comparison of the Set Median versus the Generalized Median Graph , 2009, CAIP.

[18]  Jianbo Shi,et al.  Balanced Graph Matching , 2006, NIPS.

[19]  Steven Gold,et al.  A Graduated Assignment Algorithm for Graph Matching , 1996, IEEE Trans. Pattern Anal. Mach. Intell..

[20]  Eric Mjolsness,et al.  Learning with Preknowledge: Clustering with Point and Graph Matching Distance Measures , 1996, Neural Computation.

[21]  Miquel Ferrer Sumsi.  Theory and Algorithms on the Median Graph: application to Graph-Based Classification and Clustering , 2008 .