On-Chip Optical Convolutional Neural Networks

TL;DR

This paper introduces a photonics-based circuit architecture for CNNs that significantly reduces energy consumption during inference compared to traditional electronic implementations.

Contribution

The paper presents a novel on-chip optical CNN architecture that leverages photonics to enhance energy efficiency in neural network inference.

Findings

Reduces energy consumption per inference

Demonstrates feasibility of optical CNNs on-chip

Potential for high-speed neural network processing

Abstract

Convolutional Neural Networks (CNNs) are a class of Artificial Neural Networks(ANNs) that employ the method of convolving input images with filter-kernels for object recognition and classification purposes. In this paper, we propose a photonics circuit architecture which could consume a fraction of energy per inference compared with state of the art electronics.