Meta-neural-network for Realtime and Passive Deep-learning-based Object Recognition

Abstract

Deep-learning recently show great success across disciplines yet conventionally require time-consuming computer processing or bulky-sized diffractive elements. Here we theoretically propose and experimentally demonstrate a purely-passive "meta-neural-network" with compactness and high-resolution for real-time recognizing complicated objects by analyzing acoustic scattering. We prove our meta-neural-network mimics standard neural network despite its small footprint, thanks to unique capability of its metamaterial unit cells, dubbed "meta-neurons", to produce deep-subwavelength-distribution of discrete phase shift as learnable parameters during training. The resulting device exhibits the "intelligence" to perform desired tasks with potential to address the current trade-off between reducing device's size, cost and energy consumption and increasing recognition speed and accuracy, showcased by an example of handwritten digit recognition. Our mechanism opens the route to new metamaterial-based deep-learning paradigms and enable conceptual devices such as smart transducers automatically analyzing signals, with far-reaching implications for acoustics, optics and related fields.

arXiv comment17 pages, 4 figures

References in corpus (7)

• A Survey on Deep Learning in Medical Image Analysis
• Deep Learning with Coherent Nanophotonic Circuits
• All-Optical Machine Learning Using Diffractive Deep Neural Networks
• Topological Acoustics
• A Generative Model for Inverse Design of Metamaterials
• Experimental realizations of full control of reflected waves with subwavelength acoustic metasurfaces
• Experimental Realization of Acoustic Chern Insulator

Cited by in corpus (7)

• Tailoring Structure-borne Sound Through Bandgap Engineering in Phononic Crystals and Metamaterials: A Comprehensive Review
• Deterministic and probabilistic deep learning models for inverse design of broadband acoustic cloak
• Neural network design of multilayer metamaterial for temporal differentiation
• Mechanical intelligence via fully reconfigurable elastic neuromorphic metasurfaces
• Acoustic Structure Inverse Design and Optimization Using Deep Learning
• Intelligent mechanical metamaterials towards learning static and dynamic behaviors
• Analog dual classifier via a time-modulated neuromorphic metasurface