# Digital Electronics and Analog Photonics for Convolutional Neural Networks (DEAP-CNNs)

## TL;DR

This paper introduces DEAP-CNNs, a hybrid digital-electronic and analog-photonic hardware architecture for CNNs that promises significant speed improvements while maintaining current power levels.

## Contribution

It presents a novel hybrid architecture combining digital electronics and analog photonics for CNNs, achieving up to 14 times faster processing with similar power consumption.

## Key findings

- Potential 2.8 to 14 times faster CNN processing
- Maintains comparable power usage to GPUs
- Leverages photonic analog processing advantages

## Abstract

Convolutional Neural Networks (CNNs) are powerful and highly ubiquitous tools for extracting features from large datasets for applications such as computer vision and natural language processing. However, a convolution is a computationally expensive operation in digital electronics. In contrast, neuromorphic photonic systems, which have experienced a recent surge of interest over the last few years, propose higher bandwidth and energy efficiencies for neural network training and inference. Neuromorphic photonics exploits the advantages of optical electronics, including the ease of analog processing, and busing multiple signals on a single waveguide at the speed of light. Here, we propose a Digital Electronic and Analog Photonic (DEAP) CNN hardware architecture that has potential to be 2.8 to 14 times faster while maintaining the same power usage of current state-of-the-art GPUs.