Neuromorphic silicon photonics with 50 GHz tiled matrix multiplication for deep-learning applications

The explosive volume growth of deep-learning (DL) applications has triggered an era in computing, with neuromorphic photonic platforms promising to merge ultra-high speed and energy efficiency credentials with the brain-inspired computing primitives. The transfer of deep neural networks (DNNs) onto silicon photonic (SiPho) architectures requires, however, an analog computing engine that can perform tiled matrix multiplication (TMM) at line rate to support DL applications with a large number of trainable parameters, similar to the approach followed by state-of-the-art electronic graphics processing units. Herein, we demonstrate an analog SiPho computing engine that relies on a coherent architecture and can perform optical TMM at the record-high speed of 50 GHz. Its potential to support DL applications, where the number of trainable parameters exceeds the available hardware dimensions, is highlighted through a photonic DNN that can reliably detect distributed denial-of-service attacks within a data center with a Cohen's kappa score-based accuracy of 0.636.

Automated summary

The explosive growth of deep-learning applications has led to a new era in computing, with neuromorphic photonic platforms combining high speed and energy efficiency. However, transferring deep neural networks onto silicon photonic architectures requires an analog computing engine capable of tiled matrix multiplication at line rate. In this study, we demonstrate an analog SiPho computing engine using a coherent architecture, achieving record-high optical TMM speed of 50 GHz. We highlight its potential to support DL applications with a large number of trainable parameters, showcasing a photonic DNN that can detect distributed denial-of-service attacks with high accuracy.