Analog implementation of encoded neural networks
Authors:
- 1 May 2013
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1612-1615
Abstract
Encoded neural networks mix the principles of associative memories and error-correcting decoders. Their storage capacity has been shown to be much larger than Hopfield Neural Networks'. This paper introduces an analog implementation of this new type of network. The proposed circuit has been designed for the 1V supply ST CMOS 65nm process. It consumes 1165 times less energy than a digital equivalent circuit while being 2.7 times more efficient in terms of combined speed and surface.Keywords
This publication has 8 references indexed in Scilit:
- Architecture and implementation of an associative memory using sparse clustered networks2012
- Synthesis of Bias-Scalable CMOS Analog Computational Circuits Using Margin PropagationIEEE Transactions on Circuits and Systems I: Regular Papers, 2011
- Sparse Neural Networks With Large Learning DiversityIEEE Transactions on Neural Networks, 2011
- Coded Hopfield networks2010
- An analog VLSI decoding technique for digital codes2003
- The analog decoder2002
- A CMOS analog winner-take-all network for large-scale applicationsIEEE Transactions on Circuits and Systems I: Regular Papers, 1998
- Neural networks and physical systems with emergent collective computational abilities.Proceedings of the National Academy of Sciences, 1982