tailieunhanh - neural networks algorithms applications and programming techniques phần 9

chúng ta có thể thu thập số liệu thống kê về mạng xảy ra đồng bằng cách bắt đầu tại các đơn vị đầu vào đầu tiên và tuần tự quét tất cả các đơn vị khác trong mạng. Sau khi hoàn thành này vượt qua ban đầu, chúng tôi có thể hoàn thành quá trình quét mạng bằng cách chỉ đơn thuần là incrementing con trỏ mảng của chúng tôi | ART2 317 Input vector Figure The overall structure of the ART2 network is the same as that of ART1. The Fl layer has been divided into six sublayers w x u v p and q. Each node labeled G is a gain-control unit that sends a nonspecific inhibitory signal to each unit on the layer it feeds. All sublayers on F as well as the r layer of the orienting subsystem have the same number of units. Individual sublayers on Fl are connected unit to unit that is the layers are not fully interconnected with the exception of the bottom-up connections to F2 and the top-down connections from F2. the appearance of the multiplicative factor in the first term on the right-hand side in Eq. . For the ART2 model presented here we shall set B and C identically equal to zero. As with ART1 and represent net excitatory and inhibitory factors respectively. Likewise we shall be interested in only the asymptotic solution so Jfc Xk A 1 -Ik 318 Adaptive Resonance Theory The values of the individual quantities in Eq. vary according to the sublayer being considered. For convenience we have assembled Table which shows all of the appropriate quantities for each F sublayer as well as the r layer of the orienting subsystem. Based on the table the activities on each of the six sublayers on F can be summarized by the following equations w Ii dUi Xl e ilM Vi f Xi bf qi V - Vi Ul e v Pi Ui j Th 9i p We shall discuss the orienting subsystem r layer shortly. The parameter e is typically set to a positive number considerably less than 1. It has the effect Quantity Layer A D 7 J7 w 1 1 I i au 0 x e 1 w HI u e 1 Vi IMI V 1 1 f xi bf qi 0 p 1 1 Ui y gQ 0 q e 1 Pi IIpII r c 1 u CD INI cp Table Factors in Eq. for each F sublayer and the r layer. is the jth component of the input vector. The parameters a b c and e are constants whose values will be discussed in the text yj is the activity of the jth unit on the F2 layer and g y is the output