Abstract:A novel metasurface platform is proposed and experimentally demonstrated to mimick DNNs with unprecedented capability in massively parallel processing. The designed DNNs enable functionalities in identifying digits/fashions, wavelengths, and spatial sequences between multiple targets. Unlike traditional DNNs with a single function in classifying digits/fashions, this multiplexed DNN architecture opens up new avenues for further integrating versatile functions into multichannel metasurfaces. Benefiting from low power consumption and high processing speed, there is a growing interest in diffraction neural networks (DNNs), which are typically showcased with 3D printing devices, leading to large volumes, high costs, and low levels of integration. Metasurfaces can desirably manipulate wavefronts of electromagnetic waves, providing a compact platform for mimicking DNNs with novel functions. Although multi‐wavelength and multi‐target recognition provides a richer and more detailed understanding of complex environments, existing architectures are primarily trained to classify a single target at a specific wavelength. A metasurface approach is proposed to design multiplexed DNNs that can classify multiple targets and spatial sequences across various wavelengths in multiple channels. To realize multi‐task processing, the dielectric metasurface is designed based on phase and wavelength multiplexing, which can integrate multi‐target DNNs with different tasks such as operating at distinct wavelengths and classifying diverse targets. The efficacy of this method is exemplified through the numerical simulation and experimental demonstration of recognizing a single target with two wavelengths, two targets at a single wavelength, and two targets at dual wavelengths. This compact metasurface approach enables the design of multi‐target and multi‐wavelength DNNs, opening a new window to develop massively parallel processing and versatile artificial intelligence systems.

Diffraction Deep Neural Network Based on Huygens' Metasurface for Image Classification

An Enhanced Diffractive Neural Network for Metasurface Holograms with High Resolution, Low Noise, and Uniform Intensity

Diffractive Deep Neural Network for Optical Orbital Angular Momentum Multiplexing and Demultiplexing

Diffractive Deep Neural Network for Microwave Imaging Based on a Multi-Layer Transmitarray Antenna

All-Optical Signal Processing of Vortex Beams with Diffractive Deep Neural Networks

Metasurface Enabled Multi‐Target and Multi‐Wavelength Diffraction Neural Networks

Direct electromagnetic information processing with planar diffractive neural network

Optical Analog Computing Devices Designed by Deep Neural Network

2bit Nonlinear Diffractive Deep Neural Network (2bit ND 2 NN): A quantized nonlinear all-optical diffractive deep neural network implementation

Diffractive Neural Networks with Improved Expressive Power for Grayscale Image Classification

Fourier-space Diffractive Deep Neural Network.

Diffraction Neural Network for Multi‐Source Information of Arrival Sensing

Time-lapse image classification using a diffractive neural network

Metasurface-enabled on-chip multiplexed diffractive neural networks in the visible

Advances and Progress of Diffractive Deep Neural Networks

On the Application of Deep Learning in Modeling Metasurface

Simulating an Integrated Photonic Image Classifier for Diffractive Neural Networks

Deep Learning-designed Diffractive Neural Networks

High-efficiency Transmissive Tunable Metasurfaces for Binary Cascaded Diffractive Layers

Efficient Design Optimization for Diffractive Deep Neural Networks

A programmable diffractive deep neural network based on a digital-coding metasurface array