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.

Metasurface-based Diffractive Optical Networks with Dual-channel Complex Amplitude Modulation

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

Integrated Diffractive Optical Neural Network with Space-Time Interleaving

Multi-Dimensional Reconfigurable, Physically Composable Hybrid Diffractive Optical Neural Network

Multimode Diffractive Optical Neural Network

On-chip Photonic Diffractive Optical Neural Network Based on a Spatial Domain Electromagnetic Propagation Model.

Scientific Computing with Diffractive Optical Neural Networks

On-chip 4F-System Based on Concave Mirrors for Optical Neural Networks

C-DONN: Compact Diffractive Optical Neural Network with Deep Learning Regression.

All dielectric metasurface based diffractive neural networks for 1-bit adder

High-efficiency Transmissive Tunable Metasurfaces for Binary Cascaded Diffractive Layers

Physics-aware Roughness Optimization for Diffractive Optical Neural Networks

Digitized Phase Change Material Heterostack for Diffractive Optical Neural Network

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

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

Full-Fourier-Component Tailorable Optical Neural Meta-Transformer

Physics-aware Differentiable Discrete Codesign for Diffractive Optical Neural Networks

Multiplexed Metasurfaces for Diffractive Optics via Phase Correlation Method

Parallelized and Cascadable Optical Logic Operations by Few-Layer Diffractive Optical Neural Network

Diffractive Magic Cube Network with Super-high Capacity Enabled by Mechanical Reconfiguration

Partitionable High-Efficiency Multilayer Diffractive Optical Neural Network