A Coupled CMOS Oscillator Array for 8ns and 55pJ Inference in Convolutional Neural Networks

TL;DR

This paper presents a CMOS oscillator array designed for rapid, energy-efficient inference in convolutional neural networks, achieving 8ns inference time and 55pJ energy per operation.

Contribution

It introduces a coupled CMOS oscillator array for neural network inference, demonstrating high correlation with dot products and efficient performance.

Findings

Inference time of 8ns achieved

Energy consumption of 55pJ per inference

High correlation between oscillator synchronization and dot products

Abstract

Oscillator neural networks (ONN) based on arrays of 26 CMOS ring oscillators designed and fabricated. ONN are used for inference of dot products with image fragments and kernels necessary for convolutional neural networks. The inputs are encoded as frequency shifts of oscillators using current DACs. Degree of match (DOM) is determined from oscillators synchronization. Measurements demonstrate high correlation of DOM and dot products. Inference requires the time of 8ns and energy of 55pJ.