Programmable Phase-change Metasurfaces on Waveguides for Multimode Photonic Convolutional Neural Network

Abstract

Neuromorphic photonics has recently emerged as a promising hardware accelerator, with significant potential speed and energy advantages over digital electronics, for machine learning algorithms such as neural networks of various types. Integrated photonic networks are particularly powerful in performing analog computing of matrix-vector multiplication (MVM) as they afford unparalleled speed and bandwidth density for data transmission. Incorporating nonvolatile phase-change materials in integrated photonic devices enables indispensable programming and in-memory computing capabilities for on-chip optical computing. Here, we demonstrate a multimode photonic computing core consisting of an array of programable mode converters based on metasurface made of phase-change materials. The programmable converters utilize the refractive index change of the phase-change material Ge-Sb-Te during phase transition to control the waveguide spatial modes with a very high precision of up 64 levels in modal contrast. This contrast is used to represent the matrix elements, with 6-bit resolution and both positive and negative values, to perform MVM computation in neural network algorithms. We demonstrate an optical convolutional neural network that can perform image processing and classification tasks with high accuracy. With a broad operation bandwidth and a compact device footprint, the demonstrated multimode photonic core is very promising toward a large-scale photonic processor for high-throughput optical neural networks.

17 pages, 4 figures

References in corpus (11)

• Deep Learning with Coherent Nanophotonic Circuits
• Extreme Broadband Transparent Optical Phase Change Materials for High-Performance Nonvolatile Photonics
• Mixed-Precision In-Memory Computing
• A silicon photonic modulator neuron
• Low-loss and broadband non-volatile phase-change directional coupler switches
• Nonvolatile electrically reconfigurable integrated photonic switch
• Digital Electronics and Analog Photonics for Convolutional Neural Networks (DEAP-CNNs)
• A Photonic In-Memory Computing primitive for Spiking Neural Networks using Phase-Change Materials
• Neuromorphic photonics with electro-absorption modulators
• A monolithic 56 Gb/s silicon photonic pulse-amplitude modulation transmitter
• Miniature Multi-Level Optical Memristive Switch Using Phase Change Material

Cited by in corpus (15)

• An optical neural network using less than 1 photon per multiplication
• High-order tensor flow processing using integrated photonic circuits
• In-memory photonic dot-product engine with electrically programmable weight banks
• Non-volatile electrically programmable integrated photonics with a 5-bit operation
• Broadband nonvolatile electrically programmable silicon photonic switches
• Photonic extreme learning machine by free-space optical propagation
• A Microring as a Reservoir Computing Node: Memory/Nonlinear Tasks and Effect of Input Non-ideality
• Programmable metasurfaces for future photonic artificial intelligence
• Principles and Metrics of Extreme Learning Machines Using a Highly Nonlinear Fiber
• AnalogVNN: A fully modular framework for modeling and optimizing photonic neural networks
• Selective excitation of high-order modes in two-dimensional cavity resonator integrated grating filters
• Design of an ultra-compact, energy-efficient non-volatile photonic switch based on phase change materials
• On-chip rewritable phase-change metasurface for programmable diffractive deep neural networks
• Scalable low-latency entanglement distribution for distributed quantum computing
• 2022 Roadmap on Neuromorphic Computing and Engineering