PhotoFourier: A Photonic Joint Transform Correlator-Based Neural Network Accelerator

TL;DR

PhotoFourier leverages photonic joint transform correlator technology to enable ultra-fast, energy-efficient neural network inference, significantly outperforming existing photonic accelerators in energy-delay metrics.

Contribution

It introduces a novel photonic neural network accelerator based on JTC that overcomes key challenges in on-chip photonic computing and achieves substantial performance improvements.

Findings

Over 28X better energy-delay product than current photonic accelerators

Instantaneous convolution computation via light time of flight

Addresses key challenges in on-chip photonic Fourier domain computing

Abstract

The last few years have seen a lot of work to address the challenge of low-latency and high-throughput convolutional neural network inference. Integrated photonics has the potential to dramatically accelerate neural networks because of its low-latency nature. Combined with the concept of Joint Transform Correlator (JTC), the computationally expensive convolution functions can be computed instantaneously (time of flight of light) with almost no cost. This 'free' convolution computation provides the theoretical basis of the proposed PhotoFourier JTC-based CNN accelerator. PhotoFourier addresses a myriad of challenges posed by on-chip photonic computing in the Fourier domain including 1D lenses and high-cost optoelectronic conversions. The proposed PhotoFourier accelerator achieves more than 28X better energy-delay product compared to state-of-art photonic neural network accelerators.