Your browser doesn't support javascript.
loading
Experimentally realized in situ backpropagation for deep learning in photonic neural networks.
Pai, Sunil; Sun, Zhanghao; Hughes, Tyler W; Park, Taewon; Bartlett, Ben; Williamson, Ian A D; Minkov, Momchil; Milanizadeh, Maziyar; Abebe, Nathnael; Morichetti, Francesco; Melloni, Andrea; Fan, Shanhui; Solgaard, Olav; Miller, David A B.
Afiliación
  • Pai S; Department of Electrical Engineering, Stanford University, Stanford, CA 94305, USA.
  • Sun Z; Department of Electrical Engineering, Stanford University, Stanford, CA 94305, USA.
  • Hughes TW; Department of Applied Physics, Stanford University, Stanford, CA 94305, USA.
  • Park T; Department of Electrical Engineering, Stanford University, Stanford, CA 94305, USA.
  • Bartlett B; Department of Applied Physics, Stanford University, Stanford, CA 94305, USA.
  • Williamson IAD; Department of Electrical Engineering, Stanford University, Stanford, CA 94305, USA.
  • Minkov M; Department of Electrical Engineering, Stanford University, Stanford, CA 94305, USA.
  • Milanizadeh M; Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, Milan, Italy.
  • Abebe N; Department of Electrical Engineering, Stanford University, Stanford, CA 94305, USA.
  • Morichetti F; Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, Milan, Italy.
  • Melloni A; Dipartimento di Elettronica, Informazione e Bioingegneria, Politecnico di Milano, Milan, Italy.
  • Fan S; Department of Electrical Engineering, Stanford University, Stanford, CA 94305, USA.
  • Solgaard O; Department of Electrical Engineering, Stanford University, Stanford, CA 94305, USA.
  • Miller DAB; Department of Electrical Engineering, Stanford University, Stanford, CA 94305, USA.
Science ; 380(6643): 398-404, 2023 Apr 28.
Article en En | MEDLINE | ID: mdl-37104594
Integrated photonic neural networks provide a promising platform for energy-efficient, high-throughput machine learning with extensive scientific and commercial applications. Photonic neural networks efficiently transform optically encoded inputs using Mach-Zehnder interferometer mesh networks interleaved with nonlinearities. We experimentally trained a three-layer, four-port silicon photonic neural network with programmable phase shifters and optical power monitoring to solve classification tasks using "in situ backpropagation," a photonic analog of the most popular method to train conventional neural networks. We measured backpropagated gradients for phase-shifter voltages by interfering forward- and backward-propagating light and simulated in situ backpropagation for 64-port photonic neural networks trained on MNIST image recognition given errors. All experiments performed comparably to digital simulations ([Formula: see text]94% test accuracy), and energy scaling analysis indicated a route to scalable machine learning.
Texto completo
Añadir a Mi BVS
Imprimir
XML
PubMed Links
Buscar en Google

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Idioma: En Revista: Science Año: 2023 Tipo del documento: Article País de afiliación: Estados Unidos Pais de publicación: Estados Unidos
Texto completo
Añadir a Mi BVS
Imprimir
XML
PubMed Links
Buscar en Google

Texto completo: 1 Colección: 01-internacional Base de datos: MEDLINE Idioma: En Revista: Science Año: 2023 Tipo del documento: Article País de afiliación: Estados Unidos Pais de publicación: Estados Unidos