DocumentCode :
288592
Title :
An analog VLSI neural network architecture with on-chip learning
Author :
Montalvo, Antonio J. ; Paulos, John J. ; Gyurcsik, Ronald S.
Author_Institution :
Dept. of Electr. & Comput. Eng., North Carolina State Univ., Raleigh, NC, USA
Volume :
3
fYear :
1994
fDate :
27 Jun-2 Jul 1994
Firstpage :
1364
Abstract :
A user-configurable analog VLSI feedforward neural network architecture that adds only 10% to chip area relative to a fixed topology is described. Central to the architecture is a novel synapse circuit that consumes 4500 μm2 in a 2-μm technology. Hybrid dynamic and non-volatile weight storage allows fast learning as well as reliable long-term storage. Measured synapse current-voltage curves from a test chip are presented. The synapse includes a weight increment circuit that adds offset of only 1 part in 13 bits allowing analog-domain on-chip learning. Weight update circuits that implement a semiparallel weight perturbation learning algorithm are presented
Keywords :
VLSI; analogue integrated circuits; analogue processing circuits; feedforward neural nets; learning (artificial intelligence); network topology; neural chips; neural net architecture; 2 mum; analog VLSI neural network; feedforward neural network; hybrid dynamic storage; non-volatile weight storage; on-chip learning; semiparallel weight perturbation learning; synapse circuit; synapse current-voltage curves; topology; user-configurable architecture; weight update circuits; Circuits; Computer architecture; Feedforward neural networks; Network-on-a-chip; Neural networks; Neurons; Nonvolatile memory; Semiconductor device measurement; Very large scale integration; Voltage;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Neural Networks, 1994. IEEE World Congress on Computational Intelligence., 1994 IEEE International Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-1901-X
Type :
conf
DOI :
10.1109/ICNN.1994.374484
Filename :
374484
Link To Document :
بازگشت