Programmable metasurfaces for future photonic artificial intelligence
arXiv:2505.11659 · doi:10.1038/s42254-025-00831-7
Abstract
Photonic neural networks (PNNs), which share the inherent benefits of photonic systems, such as high parallelism and low power consumption, could challenge traditional digital neural networks in terms of energy efficiency, latency, and throughput. However, producing scalable photonic artificial intelligence (AI) solutions remains challenging. To make photonic AI models viable, the scalability problem needs to be solved. Large optical AI models implemented on PNNs are only commercially feasible if the advantages of optical computation outweigh the cost of their input-output overhead. In this Perspective, we discuss how field-programmable metasurface technology may become a key hardware ingredient in achieving scalable photonic AI accelerators and how it can compete with current digital electronic technologies. Programmability or reconfigurability is a pivotal component for PNN hardware, enabling in situ training and accommodating non-stationary use cases that require fine-tuning or transfer learning. Co-integration with electronics, 3D stacking, and large-scale manufacturing of metasurfaces would significantly improve PNN scalability and functionalities. Programmable metasurfaces could address some of the current challenges that PNNs face and enable next-generation photonic AI technology.
Nat. Rev. Phys. (2025)
References in corpus (41)
- Deep Learning with Coherent Nanophotonic Circuits
- All-Optical Machine Learning Using Diffractive Deep Neural Networks
- Parallel convolution processing using an integrated photonic tensor core
- 11 TeraFLOPs per second photonic convolutional accelerator for deep learning optical neural networks
- Multiwavelength Achromatic Metasurfaces by Dispersive Phase Compensation
- Gate-tunable conducting oxide metasurfaces
- Deep physical neural networks enabled by a backpropagation algorithm for arbitrary physical systems
- Large-scale neuromorphic optoelectronic computing with a reconfigurable diffractive processing unit
- Phase-only transmissive spatial light modulator based on tunable dielectric metasurface
- Electrically Reconfigurable Nonvolatile Metasurface Using Low-Loss Optical Phase Change Material
- Tunable nanophotonics enabled by chalcogenide phase-change materials
- The physics of optical computing
- Programmable Phase-change Metasurfaces on Waveguides for Multimode Photonic Convolutional Neural Network
- Experimentally realized in situ backpropagation for deep learning in nanophotonic neural networks
- Electrical Tuning of Phase Change Antennas and Metasurfaces
- Image sensing with multilayer, nonlinear optical neural networks
- Active metasurfaces: lighting the path to commercial success
- Single chip photonic deep neural network with accelerated training
- Theory of neuromorphic computing by waves: machine learning by rogue waves, dispersive shocks, and solitons
- Gigahertz free-space electro-optic modulators based on Mie resonances
- All-Optical Information Processing Capacity of Diffractive Surfaces
- Solving integral equations in free-space with inverse-designed ultrathin optical metagratings
- Three dimensional waveguide-interconnects for scalable integration of photonic neural networks
- Electrochemically-controlled metasurfaces with high-contrast switching at visible frequencies
- The Forward-Forward Algorithm: Some Preliminary Investigations
- Compound Metaoptics for Amplitude and Phase Control of Wavefronts
- An Optical Frontend for a Convolutional Neural Network
- Why optics needs thickness
- Nonlinear optical encoding enabled by recurrent linear scattering
- Inverse-designed low-index-contrast structures on silicon photonics platform for vector-matrix multiplication
- Deep learning enabled design of complex transmission matrices for universal optical components
- A single inverse-designed photonic structure that performs parallel computing
- Reconfigurable Flat Optics with Programmable Reflection Amplitude Using Lithography-Free Phase-Change Materials Ultra Thin Films
- Cascadable all-optical NAND gates using diffractive networks
- Optical Neural Networks: The 3D connection
- Roadmap on Neuromorphic Photonics
- The hardware is the software
- Response to Comment on "All-optical machine learning using diffractive deep neural networks"
- Training of Physical Neural Networks
- The Spatial Complexity of Optical Computing and How to Reduce It
- Metasurface-based planar microlenses for SPAD pixels