Data-efficient Modeling of Optical Matrix Multipliers Using Transfer Learning

A. Cem; O. Jovanovic; S. Yan; Y. Ding; D. Zibar; F. Da Ros

Data-efficient Modeling of Optical Matrix Multipliers Using Transfer Learning

A. Cem, O. Jovanovic, S. Yan, Y. Ding, D. Zibar, F. Da Ros

Huazhong University of Science and Technology

Research output: Contribution to conference › Conference abstract for conference › Research › peer-review

Abstract

We demonstrate transfer learning-assisted neural network models for optical matrix multipliers with scarce measurement data. Our approach uses < 10% of experimental data needed for best performance and outperforms analytical models for a Mach-Zehnder interferometer mesh.

Original language	English
Publication date	2023
Number of pages	2
Publication status	Published - 2023
Event	2023 Conference on Lasers and Electro-Optics (CLEO) - San Jose McEnery Convention Center, San Jose, United States Duration: 7 May 2023 → 12 May 2023

Conference

Conference	2023 Conference on Lasers and Electro-Optics (CLEO)
Location	San Jose McEnery Convention Center
Country/Territory	United States
City	San Jose
Period	07/05/2023 → 12/05/2023

OpenUrl availability

Full text

Cite this

@conference{52947d02881f4256b2dbef109911b521,

title = "Data-efficient Modeling of Optical Matrix Multipliers Using Transfer Learning",

abstract = "We demonstrate transfer learning-assisted neural network models for optical matrix multipliers with scarce measurement data. Our approach uses < 10% of experimental data needed for best performance and outperforms analytical models for a Mach-Zehnder interferometer mesh.",

author = "A. Cem and O. Jovanovic and S. Yan and Y. Ding and D. Zibar and {Da Ros}, F.",

year = "2023",

language = "English",

pages = "1--2",

note = "2023 Conference on Lasers and Electro-Optics (CLEO) ; Conference date: 07-05-2023 Through 12-05-2023",

}

TY - ABST

T1 - Data-efficient Modeling of Optical Matrix Multipliers Using Transfer Learning

AU - Cem, A.

AU - Jovanovic, O.

AU - Yan, S.

AU - Ding, Y.

AU - Zibar, D.

AU - Da Ros, F.

PY - 2023

Y1 - 2023

N2 - We demonstrate transfer learning-assisted neural network models for optical matrix multipliers with scarce measurement data. Our approach uses < 10% of experimental data needed for best performance and outperforms analytical models for a Mach-Zehnder interferometer mesh.

AB - We demonstrate transfer learning-assisted neural network models for optical matrix multipliers with scarce measurement data. Our approach uses < 10% of experimental data needed for best performance and outperforms analytical models for a Mach-Zehnder interferometer mesh.

M3 - Conference abstract for conference

SP - 1

EP - 2

T2 - 2023 Conference on Lasers and Electro-Optics (CLEO)

Y2 - 7 May 2023 through 12 May 2023

ER -

Data-efficient Modeling of Optical Matrix Multipliers Using Transfer Learning

Abstract

Conference

OpenUrl availability

Fingerprint

Cite this