High-speed electro-optic engine for a quantized photonic neural network

Authors:
Publisher website
Google Scholar
Abstract
Integrated photonic computing possesses remarkable advantages in computational power, becoming a promising candidate for high-performance computing in the post-Moore era. However, integrated photonic computing systems face trade-offs between scale, speed, and precision, hindering their development into high-performance processors. In this work, we propose and experimentally demonstrate a photonic quantized neural network using a large-scale lithium niobate on insulator high-speed electro-optic engine. With the electro-optic engine providing 1.6 TeraOPS computational power, our photonic quantized neural network experiments achieve a high accuracy of 97% in CNN-based MNIST handwritten digit recognition and 80% in GCN-based Cora data node classification, while also realizing over 95% model compression. Based on the quantized input-weight parallel encoding scheme, our high-speed electro-optic engine achieves a 100-fold improvement in system frequency over digital computing architectures and a 10-fold improvement in both modulation speed and compute density over optoelectronic computing architectures, while also achieving competitive energy efficiency. Our experimental findings indicate that our photonic quantized neural network technology paves the way for large-scale and high-speed integrated photonic computing, further promoting its practical application.
Funding Information
  • National Key Research and Development Program of China (2024YFA1409300)
  • National Natural Science Foundation of China (62235012)
  • National Natural Science Foundation of China (12304342)
  • National Natural Science Foundation of China (12574549)
  • National Natural Science Foundation of China (12574542)
  • Innovation Program for Quantum Science and Technology (2021ZD0301500)
  • Innovation Program for Quantum Science and Technology (2021ZD0300700)
  • Science and Technology Commission of Shanghai Municipality (2019SHZDZX01)
  • Science and Technology Commission of Shanghai Municipality (24ZR1438700)
  • Science and Technology Commission of Shanghai Municipality (24ZR1430700)
  • Science and Technology Commission of Shanghai Municipality (24LZ1401500)
  • Startup Fund for Young Faculty at SJTU (24X010502876)
  • Startup Fund for Young Faculty at SJTU (24X010500170)
  • Frontier Technologies R&D Program of Jiangsu (SBF20250000094)

This publication has 59 references indexed in Scilit: