Abstract
The effects of multiplier output offsets on on-chip learning are analyzed. Offsets on signal feed-forward multipliers are equivalent to sigmoid biases and do not degrade learning performance. Offsets on multipliers for sigmoid derivatives cause static errors at the outputs, which may be overcome by proper choices of target values. Offsets on weight-adjustment multipliers cause not only output static errors but also weight drifts, which are hard to compensate. Therefore weight-adjustment multipliers are most critical for analog neuro-chips with on-chip learning capability. Simulation results agree well with analytic calculations.
Similar content being viewed by others
References
T.Morie and Y.Amemiya, “An all-analog expandable neural network LSI with on-chip backpropagation learning”, IEEE J. Solid State Circuits, Vol. 29, No. 9, pp. 1086–1093, 1994.
A.F.Murray and P.J.Edwards, “Synaptic weight noise during multilayer perceptron training: fault tolerance and training improvements”, IEEE Trans. Neural Networks, Vol. 4, No. 4, pp. 722–725, 1993.
B.K.Dolenko and H.C.Card, “Tolerance to analog hardware of on-chip learning in backpropagation networks”, IEEE Trans. Neural Networks, Vol. 6, No. 5, pp. 1045–1052, 1995.
Y.K. Choi, J.-W. Cho, S.-O. Kwon, and S.-Y. Lee, “Analog electronic neuro-chip sets with on-chip learning capability”, in Proc. Intl. Symposium on Bioelectronic and Moleculer Electronic Devices and Intl. Conf. on Molecular Electronics and Biocomputing, Okinawa, Japan, 1995, pp. 259–260.
G.W.Stewart, Introduction to Matrix Computations, Academic Press: New York, p. 220, 1973.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Choi, Y.K., Ahn, K.H. & Lee, S.Y. Effects of multiplier output offsets on on-chip learning for analog neuro-chips. Neural Process Lett 4, 1–8 (1996). https://doi.org/10.1007/BF00454840
Issue Date:
DOI: https://doi.org/10.1007/BF00454840