Title :
BP-based learning system with analog neuro-LSI
Author :
Todo, Yoshina ; Takasaki, Tsutomu ; Yoshida, Masahiro ; Yoneyama, Teru ; Asai, Hideki
Author_Institution :
Systec Corp., Shizuoka, Japan
Abstract :
This paper describes the fabrication of analog neuro-LSI and a learning system with the neuro-LSI, which uses the output of neuro-LSI for learning. First, the design and fabrication of the multi-layer neural network are shown. Next, the construction of learning system with Labview is described and the system performance is estimated. Finally, we show that this system can cancel the fluctuation of analog LSIs and is useful and practical
Keywords :
VLSI; analogue processing circuits; backpropagation; large scale integration; multilayer perceptrons; neural chips; BP-based learning system; Labview; analog LSIs; analog neuro-LSI; learning system; multi-layer neural network; system performance; Fabrication; Fluctuations; Large scale integration; Learning systems; Modeling; Multi-layer neural network; Neural networks; Neurons; Very large scale integration; Voltage;
Conference_Titel :
Circuits and Systems, 2001. MWSCAS 2001. Proceedings of the 44th IEEE 2001 Midwest Symposium on
Conference_Location :
Dayton, OH
Print_ISBN :
0-7803-7150-X
DOI :
10.1109/MWSCAS.2001.986286
