The effects of analog hardware properties on backpropagation networks with on-chip learning
- 30 December 2002
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 110-115 vol.1
- https://doi.org/10.1109/icnn.1993.298522
Abstract
Results of simulations performed assuming both forward and backward computation done on-chip using analog components are presented. Aspects of analog hardware studied are component variability (variability in multiplier gains and zero offsets), limited voltage ranges, and components (multipliers) that only approximate the computations in the backpropagation algorithm. It is shown that backpropagation networks can learn to compensate for all these shortcomings of analog circuits except for zero offsets. Variability in multiplier gains is not a problem, and learning is still possible despite limited voltage ranges and function approximations. Fixed component variation from fabrication is shown to be less detrimental to learning than component variation due to noise.Keywords
This publication has 6 references indexed in Scilit:
- VLSI design of compact and high-precision analog neural network processorsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- CMOS implementation of analog Hebbian synaptic learning circuitsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2002
- Analog CMOS implementation of a multilayer perceptron with nonlinear synapsesIEEE Transactions on Neural Networks, 1992
- Back-propagation learning and nonidealities in analog neural network hardwareIEEE Transactions on Neural Networks, 1991
- Statistical pattern recognition with neural networks: benchmarking studiesPublished by Institute of Electrical and Electronics Engineers (IEEE) ,1988
- A high-performance monolithic multiplier using active feedbackIEEE Journal of Solid-State Circuits, 1974