Convolution Inference via Synchronization of a Coupled CMOS Oscillator Array

IEEE Journal on Exploratory Solid-State Computational Devices and Circuits

Hyung-Jin Lee¹  Yongping Fan¹  James S. Ayers¹  Peter Kurahashi¹  Telesphor Kamgaing²  Georgios C. Dogiamis²  Hai Li³  I. A. Young³  Dmitri E. Nikonov³ 

[1] Advanced Design, Intel Corporation, Hillsboro, OR, USA;Components Research, Intel Corporation, Chandler, AZ, USA;Components Research, Intel Corporation, Hillsboro, OR, USA;

关键词: CMOS ring oscillators;    convolution;    coupled oscillators;    neural networks (NNs);    synchronization;   

DOI  :  10.1109/JXCDC.2020.3046143

来源: DOAJ

【 摘 要 】

Oscillator neural networks (ONNs) are a promising hardware option for artificial intelligence. With an abundance of theoretical treatments of ONNs, few experimental implementations exist to date. In contrast to prior publications of only building block functionality, we report a practical experimental demonstration of neural computing using an ONN. The arrays contain 26 CMOS ring oscillators in the GHz range of frequencies are tuned by image data and convolution kernels. Synchronization of oscillators results in an analog output voltage approximating convolution neural network operation.

【 授权许可】

Unknown