Abstract
In this paper we present a VLSI systolic architecture for implementing the multilayer perceptron network. Both the retrieving phase and the backpropagation learning algorithm with a sigmoid nonlinearity are considered. The implementation is based upon a variable size ring systolic array which is highly regular thus favouring integration. This architecture can also be expanded in a parallel or cascade fashion to implement arbitrary size networks.
Preview
Unable to display preview. Download preview PDF.
References
R. Hecht-Nielsen. "Neurocomputing: picking the human brain". IEEE Spectrum, 36–41, March, 1988.
K. Karna & D. M. Breen. "An artificial neural networks tutorial: part 1 — basics". Neural Networks, Vol. 1, 4–23, January, 1989.
K. Goser. "VLSI technologies for artificial neural networks". IEEE MICRO, 28–44, December, 1989.
H. Yoon, J. H. Nang, and S. R. Maeng. "Parallel Simulation of Multilayered Neural Networks on Distributed-Memory Multiprocessors". Microproc. & Microprogramming, 29, 185–195, 1990.
L. Atlas & Y Suzuki. "Digital systems for artificial neural networks". IEEE circuits and devices magazine, 20–24, November, 1989.
M. Witbrock & M. Zagha. "An implementation of backpropagation learning on GF11, a large SIMD parallel computing". Parallel Computing, 14, 329–346, 1990.
X. Zhang et al. "The backpropagation algorithm on grid and hypercube architectures". Parallel computing, 14, 317–327, 1990.
P. Bofill et al "A systolic algorithm for backpropagation: Mapping onto a transputer network". Proceedings 2nd International Conference of the OUG Artificial Intelligence SIG. IOS Press, 1990.
J. R. Millan & P. Bofill. "Learning by backpropagation: a systolic algorithm and its transputer implementation". Neural Networks, Vol. 1, #3, 119–137, July, 1989.
A. H. Mascia and R. Ishii. "Neural Net Implementation on Single-Chip Digital Signal Processor". IECON 89, Philadelphia, PA, USA, Vol. 4, 764–769, November, 1989.
F. Blayo & P. Hurat. "A VLSI systolic array dedicated to Hopfiel neural network". VLSI for artificial intelligence. Kluwer Academic Publishers, 255–264, 1989.
A. Masaki, Y. Hirai and M. Yamada. "Neural Networks in CMOS: a case study". Circuits and Devices, 12–17, July, 1990.
M. Weinfield. "A fully digital integrated CMOS Hopfield Network including the learning algorithm". In: VLSI for Artificial Intelligence, Kluwer Academic Publishers, 169–178, 1989.
C. Mead. "Analog VLSI and Neural Networks". Addison Wesley, Reading, Mass., 1988.
M. Walker, P. Hasler and L. Akers. "A CMOS Neural Network for Pattern Association". IEEE Micro, 68–74, October, 1989.
J. Alspector. "Neural-Style microsystems that learn". IEEE Com. Magazine, 29–36, November, 1989.
L. Personnaz; A. Johannet & G. Dreyfus. "Problems and trends in integrated neural networks". Connectionism in perspective. Elsevier Science publishers B.V. (North Holland), 499–505, 1989.
J. N. Hwang; J. A. Vlontzos & S.Y. Kung. "A systolic neural network for hidden Markov Models". IEEE transactions on acoustics, speech and signal processing, Vol. 37, #12, 1967–1979, 1989.
S. Y. Kung. "Parallel architectures for artificial neural nets". Int. Conf. on Systolic Arrays, San Diego, CA, USA, 163–174, May, 1988.
S. Y. Kung & J. N. Hwang. "A unified systolic architecture for artificial neural networks". Journal of Parallel an Distributed Computing, 6, 358–387, 1989.
J. M. Moreno, F. Castillo & J. Cabestany. "Implementación VLSI de un procesador neuronal". Comunicaciones V Esc. Microelectronica, 329–332, September, 1990. In Spanish.
D. J. Burr. "Experiments on Neural Net Recognition of Spoken and Written Text". IEEE trans. on Acoust. Speech & Signal Proc., Vol. 36, #7, 1162–1168, 1988.
J. C. Lupo. "Defense Applications of Neural Networks". IEEE Com. Magazine, 82–88, Nov., 1989.
S. Kung, "VLSI Array Processors for Sygnal/Image Processing". New Comp. Tech., C-15, 561–608.
J. J, Navarro, J. M Llaberia, and M. Valero. "Partitioning: an essential step in mapping algorithms into systolic array processors". IEEE computing, 20, 77–89, July, 1987.
M. C. Chen. A Design Methodology for Synthesizing Parallel Algorithms and Architectures". Journal of Parallel and Distributed Computing, 3, 461–491, April, 1986.
S. Manohar & G. Baudet. "Pragmatic approach to systolic design". IEEE proceedings, Vol. 137, #4, 277–282, 1990.
D. I. Moldovan & J. A. B. Fortes. "Partitioning and mapping algorithms into fixed size systolic arrays". IEEE transactions on computers, Vol. c-35, #1, 1–12, 1986.
M. Valero-García; J. J. Navarro; J. M. Llaberia & V. Valero. "Systematic Hardware adaptation for systolic algoriyhms". ACM, 96–104, 1989.
D. E. Rumelhart, G. E. Hinton and R. J. Williams. "Learning Internal Representations By Error Correction". In: Parallel Distributed Processing: Explorations in the Microstructures of Cognition, Vol. 1, Cambridge, MA: MIT Press, 318–362, 1986.
R. Lippmann. "An introduction to computing with neural nets". IEEE ASSP Mag., 4–22, April, 1987.
D. J. Myers & R. A. Hutchinson. "Efficient implementation of piecewise linear activation function for digital neural networks". Electronic letters, Vol., 25, #24, 1662–1663, 1989.
S. Barro and A. Yañez. "A VLSI and WSI systolic architecture for multilayer feedforward networks of arbitrary size". IJCNN-91, Seattle, July, 1991.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1991 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Yáñez, A., Barro, S., Bugarin, A. (1991). Backpropagation multilayer perceptron: A modular implementation. In: Prieto, A. (eds) Artificial Neural Networks. IWANN 1991. Lecture Notes in Computer Science, vol 540. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0035905
Download citation
DOI: https://doi.org/10.1007/BFb0035905
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-54537-8
Online ISBN: 978-3-540-38460-1
eBook Packages: Springer Book Archive