Non-volatile Reconfigurable Digital Optical Diffractive Neural Network Based on Phase Change Material

2023/05/18 by Chu Wu, Jingyu Zhao, Wu, Chu +7
Computer Science · Engineering · Materials Science · #Advanced Memory and Neural Computing #Emerging Technologies (cs.ET) #FOS: Computer and information sciences #FOS: Electrical engineering #FOS: Physical sciences #Neural Networks and Reservoir Computing #Optics (physics.optics) #Phase-change materials and chalcogenides #Signal Processing (eess.SP) #electronic engineering #information engineering

paper · pdf · doi:10.48550/arxiv.2305.11196

openalex publication_date 2023/05/18 · openalex created_date 2023/05/23 · openalex updated_date 2026/07/28

Abstract

Optical diffractive neural networks have triggered extensive research with their low power consumption and high speed in image processing. In this work, we propose a reconfigurable digital all-optical diffractive neural network (R-ODNN) structure. The optical neurons are built with Sb2Se3 phase-change material, making our network reconfigurable, digital, and non-volatile. Using three digital diffractive layers with 14,400 neurons on each and 10 photodetectors connected to a resistor network, our model achieves 94.46% accuracy for handwritten digit recognition. We also performed full-vector simulations and discussed the impact of errors to demonstrate the feasibility and robustness of the R-ODNN.

Related

• On-chip rewritable phase-change metasurface for programmable diffractive deep neural networks
• Digitized Phase Change Material Heterostack for Diffractive Optical Neural Network
• 1-bit Quantized On-chip Hybrid Diffraction Neural Network Enabled by Authentic All-optical Fully-connected Architecture
• All-Optical Machine Learning Using Diffractive Deep Neural Networks
• Multi-Dimensional Reconfigurable, Physically Composable Hybrid Diffractive Optical Neural Network