Co-Design of Free-Space Metasurface Optical Neuromorphic Classifiers for High Performance

TL;DR

This paper investigates free-space metasurface optical classifiers that perform scene feature classification directly through light diffraction, achieving high accuracy with fewer features and highlighting the importance of co-design in system architecture, material structure, and input light.

Contribution

It introduces a theoretical and computational framework for designing metasurface-based optical classifiers that outperform traditional linear classifiers in accuracy and feature efficiency.

Findings

Single layer metasurfaces achieve ~96% accuracy on MNIST.

Fewer diffractive features are needed compared to previous methods.

Performance scales mainly with the number of apertures, suggesting benefits of multi-layer designs.

Abstract

Classification of features in a scene typically requires conversion of the incoming photonic field into the electronic domain. Recently, an alternative approach has emerged whereby passive structured materials can perform classification tasks by directly using free-space propagation and diffraction of light. In this manuscript, we present a theoretical and computational study of such systems and establish the basic features that govern their performance. We show that system architecture, material structure, and input light field are intertwined and need to be co-designed to maximize classification accuracy. Our simulations show that a single layer metasurface can achieve classification accuracy better than conventional linear classifiers, with an order of magnitude fewer diffractive features than previously reported. For a wavelength {\lambda}, single layer metasurfaces of size with aperture density achieve ~96% testing accuracy on the MNIST dataset, for an optimized distance ~ to the output plane. This is enabled by an intrinsic nonlinearity in photodetection, despite the use of linear optical metamaterials. Furthermore, we find that once the system is optimized, the number of diffractive features is the main determinant of classification performance. The slow asymptotic scaling with the number of apertures suggests a reason why such systems may benefit from multiple layer designs. Finally, we show a trade-off between the number of apertures and fabrication noise.