Orthodoxy Basis Functions and Convergence Property in Procedure Neural Networks

Learning algorithm and application of segment procedure neural network mode

“…The selection of ui(t) in the formula (7) can refer to the reference [5], just take selection of the Chebshov orthodoxy polynomial as following.…”

“…Ref [4] discusses two kinds of neural networks in details which are basis-expanded procedure neural network and projected-expanded procedure neural network. Ref [5] proposes the procedure neural network containing normal basis function. In order to control the intermediate result of the procedure or to enforce subsection object programming, Ref [6] presents a model which takes the procedure into different subsections for individual consideration based on the procedure neural network above.…”

Segment procedure neural networks

“…Paper [5] brings the feedback procedure neural networks. Papers [6,7] propose the procedure neural network containing normal basis function. In order to control the intermediate result of the procedure or to enforce segment object programming, this paper presents a model, which takes the procedure into segment for each consideration based on the procedure neural networks above.…”

“…Assume that the desire output of yi is "i (i=1,2,...,m) consider the form of the square error between the factual output and due output, that is The selection of u1(t) in the formula (11) can refer to the reference [7], just as the selection of the Chebshov orthodoxy polynomial.…”

“…Since the appearance ofthe procedure neural network [1], many literatures have researched on the problems [2,3,4,5,6,7,8]. This paper focuses on the segment model of procedure neural networks, and is composed of as following.…”