Title :
Convergence of Completely Connected Recursive Neural Networks
Author :
Zhujun, Fan ; Min, Fan
Author_Institution :
Sch. of Math. & Comput. Sci., Sun Yat-sen Univ., Guangzhou
Abstract :
This paper discusses the convergence of completely connected recursive neural networks. That is, we analyze the affection of output of the neural networks under which the error donpsilat accumulate, brought by disturbances of inputs and weights. The corresponding conditions are obtained, which are also the conditions of convergence of the neural networks.
Keywords :
neural nets; convergence; recursive neural network; Arithmetic; Computer networks; Computer science; Convergence; Impedance matching; Mathematics; Neural networks; Software engineering; Sun; Supervised learning; convergence; recursive neural networks;
Conference_Titel :
Computer Science and Software Engineering, 2008 International Conference on
Conference_Location :
Wuhan, Hubei
Print_ISBN :
978-0-7695-3336-0
DOI :
10.1109/CSSE.2008.283