Experimentally realized in situ backpropagation for deep learning in photonic neural networks

• Published in: Science (American Association for the Advancement of Science) Vol. 380; no. 6643; pp. 398 - 404
• Main Authors: 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
• Format: Journal Article
• Language: English
• Published: United States The American Association for the Advancement of Science 28-04-2023
• Subjects: Alternative energy sources; Back propagation networks; Deep learning; Energy costs; Energy efficiency; Machine learning; Neural networks; Photonics; Training; Waveguides
• ISSN: 0036-8075; 1095-9203
• DOI: 10.1126/science.ade8450

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.