Title :
On the sample mean of graphs
Author :
Jain, Brijnesh ; Obermayer, Klaus
Abstract :
We present an analytic and geometric view of the sample mean of graphs. The theoretical framework yields efficient subgradient methods for approximating a structural mean and a simple plug-in mechanism to extend existing central clustering algorithms to graphs. Experiments in clustering protein structures show the benefits of the proposed theory.
Keywords :
approximation theory; gradient methods; graph theory; learning (artificial intelligence); pattern clustering; sampling methods; central graph clustering algorithm; competitive learning; geometric view; graph sample mean; incremental mean algorithm; plug-in mechanism; protein structure clustering; statistical pattern recognition; structural mean approximation; subgradient method; Bridges; Clustering algorithms; Cost function; Euclidean distance; Frequency measurement; Machine learning; Machine learning algorithms; Pattern recognition; Principal component analysis; Proteins;
Conference_Titel :
Neural Networks, 2008. IJCNN 2008. (IEEE World Congress on Computational Intelligence). IEEE International Joint Conference on
Conference_Location :
Hong Kong
Print_ISBN :
978-1-4244-1820-6
Electronic_ISBN :
1098-7576
DOI :
10.1109/IJCNN.2008.4633920
