DocumentCode :
393852
Title :
A compressed implementation of neural network with finite physical quantities on FPGAs
Author :
Sotohebo, Takashi ; Watanabe, Minoru ; Kobayashi, Fuminori
Author_Institution :
Kyushu Inst. of Technol., Japan
Volume :
3
fYear :
2002
fDate :
5-7 Aug. 2002
Firstpage :
1861
Abstract :
This paper presents the implementation of a multilayer neural network with finite physical quantities on a FPGA. The FPGA implementation extensively reduces much computer time for the neural network that requires iterative operation to exchange the physical quantities between neurons little by little.
Keywords :
field programmable gate arrays; neural nets; self-organising feature maps; FPGAs; compressed implementation; finite physical quantities; iterative operation; neural network; Circuits; Computer networks; Digital images; Equations; Field programmable gate arrays; Iterative algorithms; Multi-layer neural network; Neural networks; Neurons; Physics computing;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
SICE 2002. Proceedings of the 41st SICE Annual Conference
Print_ISBN :
0-7803-7631-5
Type :
conf
DOI :
10.1109/SICE.2002.1196606
Filename :
1196606
Link To Document :
بازگشت