A0954
Title: Functional-edged network modeling
Authors: Chen Zhang - Tsinghua University (China) [presenting]
Abstract: Contrasting with existing works, which all consider nodes to be functions and use edges to represent the relationships between different functions, the target is network modelling, whose edges are functional data and transform the adjacency matrix into a functional adjacency tensor, introducing an additional dimension dedicated to function representation. Tucker functional decomposition is used for the functional adjacency tensor, and to further consider the community between nodes, the basis matrices are regularized to be symmetrical. Furthermore, to deal with irregular observations of the functional edges, model inference is conducted to solve a tensor completion problem. It is optimized by a Riemann conjugate gradient descent method. Besides these, several theorems are also derived to show the desirable properties of the functional edged network model. Finally, the efficacy of the proposed model is evaluated using simulation data and real metro system data from Hong Kong and Singapore.
