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. © 2010 Springer-Verlag Berlin Heidelberg.
CITATION STYLE
Jain, B. J., & Obermayer, K. (2010). Large sample statistics in the domain of graphs. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6218 LNCS, pp. 690–697). https://doi.org/10.1007/978-3-642-14980-1_68
Mendeley helps you to discover research relevant for your work.