Nonlinear Dimensionality Reduction for Low Data Regimes in Photonics Design

Efficient exploration of high-dimensional parameter space is essential in modern photonic component design. Linear dimensionality reduction such as principal component analysis has proven useful in identifying lower dimensional subspace of interest in several design problems. Yet such subspaces ofte...

Full description

Saved in:
Bibliographic Details
Published in2022 Photonics North (PN) p. 1
Main Authors Grinberg, Yuri, Al-Digeil, Muhammad, Kamandar Dezfouli, Mohsen, Melati, Daniele, Schmid, Jens H., Cheben, Pavel, Janz, Siegfried, Xu, Danxia
Format Conference Proceeding
LanguageEnglish
Published IEEE 24.05.2022
Subjects
autoencoder
Dimensionality reduction
machine learning
nanophotonic design
Neural networks
Photonics
Principal component analysis
silicon photonics
small data
Online AccessGet full text

Cover

More Information
Summary:Efficient exploration of high-dimensional parameter space is essential in modern photonic component design. Linear dimensionality reduction such as principal component analysis has proven useful in identifying lower dimensional subspace of interest in several design problems. Yet such subspaces often exhibit curvature reflecting nonlinear relationships between design parameters. For such systems linear dimensionality reduction methods can be suboptimal. We discuss how an appropriate architecture for an autoencoder neural network along with a numerically robust initialization, show improved performance compared to linear methods even in low data regimes, which are typical for photonic design problems.
ISSN:2693-8316
DOI:10.1109/PN56061.2022.9908251