Uncertain network reasoning for agents via Gaussian radial basis functions

Networking exits in either natural science or social science. There is a lot of information hidden in a network, when one tries to understand how a network influences some observed output data. How could one find out the relation between the network and the output data becomes an important issue. In this paper, I utilize radial basis functions (RBF) to analyze and extract information from such network. RBF is a good technique for interpolation of some data via a radial basis. This basis represents the basic constituent of a network. Unlike the typical settings, in this paper, I devise some algorithms in finding the best-match input radial basis and the output under uncertainty in which the cause and effect between the network and the output is not predetermined, but some kind of learning process is needed to create such relation.