Training of Photonic Neural Networks through In Situ Backpropagation
By Tyler W. Hughes, Momchil Minkov, Yu Shi, Shanhui Fan. Conference on Lasers and Electro-Optics, 2018
440 citations recorded by the source corpus.
We provide a protocol for training photonic neural networks based on adjoint methods. The gradient of the network with respect to its tunable degrees of freedom is computed by physically backpropagating an optical error signal. © 2019 The Author(s)
Subfields: Artificial Intelligence, Electrical and Electronic Engineering, Statistical and Nonlinear Physics
View Paper: https://www.semanticscholar.org/paper/2e7c6ae337127c883cba8beae03a5cb2b79eb798
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Shanhui Fan
- Rohit Unni
- Francesco Morichetti
- Kan Yao
- Andrea Melloni
- Tyler W. Hughes
- Yuebing Zheng
- Yichen Shen
In Electrical and Electronic Engineering
- Francesco Morichetti
- Olav Solgaard
- Shanhui Fan
- Rohit Unni
- Kan Yao
- Andrea Melloni
- Tyler W. Hughes
- Yuebing Zheng
In Statistical and Nonlinear Physics
Citations in the corpus
- Photonic matrix multiplication lights up photonic accelerator and beyond (2022, methodology): “12b, the input images are divided into small patches, and these patches are converted into small matrices for MVM operations.”
- Intelligent nanophotonics: merging photonics and artificial intelligence at the nanoscale (2018, background): “developed an alternative protocol in which the on-chip training is accomplished only by in situ intensity measurements [163].”
- Experimentally realized in situ backpropagation for deep learning in photonic neural networks (2022, methodology): “[5] and used in our main backpropagation training demonstration.”