Meta-optic Accelerators for Object Classifiers

TL;DR

This paper introduces a meta-optic neural network accelerator that leverages metasurfaces to perform convolution operations optically, significantly reducing energy consumption and enabling high-speed, low-power object classification.

Contribution

The work presents a novel meta-optic based neural network architecture that off-loads convolutional computations into optics, enhancing efficiency and enabling integrated optical-digital classification.

Findings

Achieved 95% accuracy in handwriting digit classification.

Attained 94% accuracy in joint digit and polarization classification.

Demonstrated robust, low-power optical neural network performance.

Abstract

Rapid advances in deep learning have led to paradigm shifts in a number of fields, from medical image analysis to autonomous systems. These advances, however, have resulted in digital neural networks with large computational requirements, resulting in high energy consumption and limitations in real-time decision making when computation resources are limited. Here, we demonstrate a meta-optic based neural network accelerator that can off-load computationally expensive convolution operations into high-speed and low-power optics. In this architecture, metasurfaces enable both spatial multiplexing and additional information channels, such as polarization, in object classification. End-to-end design is used to co-optimize the optical and digital systems resulting in a robust classifier that achieves 95% accurate classification of handwriting digits and 94% accuracy in classifying both the digit and its polarization state. This approach could enable compact, high-speed, and low-power image and information processing systems for a wide range of applications in machine-vision and artificial intelligence.