Functional networks and applications: A survey

摘要: Abstract Functional networks (FNs) are extensions of neural (NNs). Unlike NNs, FNs considers general functional models instead sigmoid-like models. Additionally, in FNs, there no weights associated with the links that connect neurons. In this paper, we review research progress and applications recent years. First, introduce architecture, three typical learning process, explain differences between NNs FNs. Second, discuss have been introduced many fields, such as time series prediction, differential equations, pattern classification, detection approximation computation, complex system modeling, computer-aided design (CAD), linear nonlinear regression. Finally, present some remarks on future directions for