DocumentCode
2389428
Title
SPIN: a sequential pipelined neurocomputer
Author
Vassiliadis, S. ; Pechanek, G.G. ; Delgado-Frias, J.G.
Author_Institution
IBM, Endicott, NY, USA
fYear
1991
fDate
10-13 Nov 1991
Firstpage
74
Lastpage
81
Abstract
A novel digital network architecture, the sequential pipelined neurocomputer (SPIN), is proposed. The SPIN processor emulates neural networks, producing high performance with minimal hardware by sequentially processing each neuron in the modeled completely connected network with a pipelined physical neuron structure. In addition to describing SPIN, performance equations are estimated for the ring systolic, the recurrent systolic array, and the neuromimetic neurocomputer architectures, three previously reported schemes for the emulation of neural networks, and a comparison with the SPIN architecture is reported
Keywords
neural nets; parallel processing; pipeline processing; systolic arrays; SPIN processor; digital network architecture; neural networks; neuromimetic neurocomputer architectures; recurrent systolic array; ring systolic array; sequential pipelined neurocomputer; Computer architecture; Costs; Emulation; Equations; Neural network hardware; Neural networks; Neurons; Recurrent neural networks; Systolic arrays; Tree data structures;
fLanguage
English
Publisher
ieee
Conference_Titel
Tools for Artificial Intelligence, 1991. TAI '91., Third International Conference on
Conference_Location
San Jose, CA
Print_ISBN
0-8186-2300-4
Type
conf
DOI
10.1109/TAI.1991.167078
Filename
167078
Link To Document