Highly Reliable Inference System of Neural Networks Using Gated Schottky Diodes

Suhwan Lim,Byung-Gook Park,Sung-Tae Lee,Jong-Ho Lee,Jong-Ho Bae,Chul-Heung Kim,Dongseok Kwon,Jai-Ho Eum,Hyeongsu Kim

doi:10.1109/jeds.2019.2913146

Suhwan Lim, Byung-Gook Park + Show 7 more

Open Access

https://doi.org/10.1109/jeds.2019.2913146

Copy DOI

Abstract

An inference system using gated Schottky diode (GSD) is proposed for highly reliable hardware-based neural networks (HNNs). We explain the characteristics of the GSD and present circuits that take into account the characteristics of the device. The reverse current of the GSD, which is the synaptic current, is saturated with respect to input voltage, which results in immunity of input and output noise and overcoming the IR drop problem in metal wire. In order to take advantages of this saturated $I-V$ characteristics, pulse-width modulation (PWM) of input data instead of amplitude modulation is proposed. In addition, by applying identical pulses to the bottom gate, the synaptic current of the GSD increases linearly, which makes it easy to transfer the calculated weights to the conductance of GSDs. By considering these characteristics, electronic circuits for PWM, current sum, and activation function are designed. Through SPICE simulation, we evaluate the inference accuracy of a 2-layer neural network. The classification accuracy rate of 100 images of MNIST test sets is 94% accuracy obtained with software.

Highlights

Neuromorphic computing using electronic synaptic devices, which is hardware-based neural networks (HNNs), is under widespread research due to its capability of low power and massively parallel operations [1]
We focus on the offchip training using gated Schottky diodes (GSD)
PULSE-WIDTH MODULATION (PWM) FOR INPUT BIAS It is commonly assumed that the I-V characteristic of an electronic synaptic device is linear in a hardware-based neural network composed of electronic synaptic devices [19]

Summary

Introduction

Neuromorphic computing using electronic synaptic devices, which is hardware-based neural networks (HNNs), is under widespread research due to its capability of low power and massively parallel operations [1]. GATED SCHOTTKY DIODE (GSD) When performing the forward propagation of neural networks using electronic synaptic devices, the characteristics of synaptic device should be considered. B. PULSE-WIDTH MODULATION (PWM) FOR INPUT BIAS It is commonly assumed that the I-V characteristic of an electronic synaptic device is linear in a hardware-based neural network composed of electronic synaptic devices [19].

Results

Conclusion