High-clockrate free-space optical in-memory computing

TL;DR

A new optical computing system uses lasers and light modulators to perform fast, energy-efficient neural network computations for edge devices.

Contribution

The paper introduces FAST-ONN, a high-clockrate free-space optical in-memory computing system for efficient deep learning.

Findings

FAST-ONN achieves billions of optical convolutions per second with ultralow latency and power consumption.

The system enables accurate inference with signed weight values using parallel differential readout in a single shot.

VCSEL transmitters can improve clockrates to over gigahertz, enabling scalable free-space optical computing hardware.

Abstract

The ability to process and act on data in real time is increasingly critical for applications ranging from autonomous vehicles, three-dimensional environmental sensing, and remote robotics. However, the deployment of deep neural networks (DNNs) in edge devices is hindered by the lack of energy-efficient scalable computing hardware. Here, we introduce a fanout spatial time-of-flight optical neural network (FAST-ONN) that calculates billions of convolutions per second with ultralow latency and power consumption. This is enabled by the combination of high-speed dense arrays of vertical-cavity surface-emitting lasers (VCSELs) for input modulation with spatial light modulators of high pixel counts for in-memory weighting. In a three-dimensional optical system, parallel differential readout allows signed weight values for accurate inference in a single shot. The performance is benchmarked with feature extraction in You-Only-Look-Once (YOLO) for convolution at 100 million frames per second (MFPS), and in-system backward propagation training with photonic reprogrammability. The VCSEL transmitters are implementable in any free-space optical computing systems to improve the clockrate to over gigahertz, where the high scalability in device counts and channel parallelism enables a new avenue to scale up free space computing hardware.

We demonstrated high-speed VCSEL in-memory neural networks that deliver billion optical convolutions per second for massively parallel edge intelligence at ultralow energy and latency.

Figures