Parallel Programming of an Arbitrary Feedforward Photonic Network

Reconfigurable photonic mesh networks of tunable beamsplitter nodes can linearly transform <inline-formula><tex-math notation="LaTeX">N</tex-math></inline-formula>-dimensional vectors representing input modal amplitudes of light for applications such as energy-effic...

Full description

Saved in:
Bibliographic Details
Published in:IEEE journal of selected topics in quantum electronics Vol. 26; no. 5; pp. 1 - 13
Main Authors: Pai, Sunil, Williamson, Ian A. D., Hughes, Tyler W., Minkov, Momchil, Solgaard, Olav, Fan, Shanhui, Miller, David A. B.
Format: Journal Article
Language:English
Published: New York IEEE 01.09.2020
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Subjects:
Adaptive optics
Computer simulation
Data processing
Demultiplexing
Finite element method
graph theory
Handwriting
Image classification
Machine learning
Neural networks
neuromorphic computing
Nodes
Optical computing
Optical fiber networks
optical neural networks
Optical thickness
Optoelectronics
Parallel programming
Phase shifters
Photonics
Programmable optical networks
Quantum phenomena
Transmission line matrix methods
Tuning
ISSN:1077-260X, 1558-4542
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Reconfigurable photonic mesh networks of tunable beamsplitter nodes can linearly transform <inline-formula><tex-math notation="LaTeX">N</tex-math></inline-formula>-dimensional vectors representing input modal amplitudes of light for applications such as energy-efficient machine learning hardware, quantum information processing, and mode demultiplexing. Such photonic meshes are typically programmed and/or calibrated by tuning or characterizing each beam splitter one-by-one, which can be time-consuming and can limit scaling to larger meshes. Here we introduce a graph-topological approach that defines the general class of feedforward networks and identifies columns of non-interacting nodes that can be adjusted simultaneously. By virtue of this approach, we can calculate the necessary input vectors to program entire columns of nodes in parallel by simultaneously nullifying the power in one output of each node via optoelectronic feedback onto adjustable phase shifters or couplers. This parallel nullification approach is robust to fabrication errors, requiring no prior knowledge or calibration of node parameters and reducing programming time by a factor of order <inline-formula><tex-math notation="LaTeX">N</tex-math></inline-formula> to being proportional to the optical depth (number of node columns). As a demonstration, we simulate our programming protocol on a feedforward optical neural network model trained to classify handwritten digit images from the MNIST dataset with up to 98% validation accuracy.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1077-260X
1558-4542
DOI:10.1109/JSTQE.2020.2997849