Pré-visualização3 páginas / 734
Pesquisar no resumo do documento
R [1 E [ã E da HE ES) Neural Network Design Thisbock provides a clear and detailed survey of basic neural network architectures and learning rules. lt, the authors emphasize mathematical analysis of networks, methods or training networks, and application of networks to practical engineering problems in pattem recognitior, signal procesing, and control systems. Features: +Extensive coverage of performance learning, including the Widrowe-Hoff rule, back propagation, and several enhancements of back propagation (e, conjugate gradient, Levenberg-Marcquardt variations) «Discussion of recurrentassociative memory networks (eg, Hopfied network «Detailed examples and mumerous solved problems +Associative and competitive networks (including feature maps, learning vector quanti- zation, and adaptive resonance theory) are explained using simple building blocks. Demonstrations on bound-in disk using MATLAB 4.0 (student | have rarely reviewed 1 WOry stronghy that this is 15 this well written. The dl a the demonstrations really support one's intultion — Professor Stan Ahalt, Olio State University n excollont ba T pl d greatiy to the [Ban nitosara 5: wwwchina-pub com AKFAMENAERNIS 10007 | ad DOOGNOO2M aves (010) dsosaso BIG cridostregputon.com Ter ui oBa Ta | BEBA | ISBNT-HI-OBA LTP - 2589 ih: 60007 + ques 4 PO SE SIT (RISCAR) a: «€ x SÊ é “om o . - £O [a = SÉ -L SS « zo o z . 8. 2 Duo DD.