Single-shot optical neural network

Abstract

Analog optical and electronic hardware has emerged as a promising alternative to digital electronics to improve the efficiency of deep neural networks (DNNs). However, previous work has been limited in scalability (input vector length K ≈ 100 elements) or has required nonstandard DNN models and retraining, hindering widespread adoption. Here, we present an analog, CMOS–compatible DNN processor that uses free-space optics to reconfigurably distribute an input vector and optoelectronics for static, updatable weighting and the nonlinearity—with K ≈ 1000 and beyond. We demonstrate single-shot-per-layer classification of the MNIST, Fashion-MNIST, and QuickDraw datasets with standard fully connected DNNs, achieving respective accuracies of 95.6, 83.3, and 79.0% without preprocessing or retraining. We also experimentally determine the fundamental upper bound on throughput (∼0.9 exaMAC/s), set by the maximum optical bandwidth before substantial increase in error. Our combination of wide spectral and spatial bandwidths enables highly efficient computing for next-generation DNNs.

Document Details

Document Type
Pub Defense Publication
Publication Date
Jun 23, 2023
Source ID
10.1126/sciadv.adg7904

Entities

People

  • Alexander Sludds
  • Christopher Panuski
  • Dirk Englund
  • Liane Bernstein
  • Ryan Hamerly
  • Sivan Trajtenberg-mills

Organizations

  • Massachusetts Institute of Technology

Tags

Fields of Study

  • Computer science
  • Physics

Readers

  • Integrated Circuit Design and Technology.
  • Neural Network Machine Learning.
  • Quantum Dot Semiconductor Device Photonics and Graphene Optoelectronic Materials and THz Physics.

Technology Areas

  • AI & ML
  • AI & ML - Neural Networks
  • Microelectronics
  • Space
  • Space - Space Objects