Experimental Demonstration of Imperfection-Agnostic Local Learning Rules on Photonic Neural Networks with Mach-Zehnder Interferometric Meshes

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Srouji, Luis El, On, Mehmet Berkay, Lee, Yun-Jhu, Abdelghany, Mahmoud, Yoo, S. J. Ben
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909064486715392
author Srouji, Luis El
On, Mehmet Berkay
Lee, Yun-Jhu
Abdelghany, Mahmoud
Yoo, S. J. Ben
author_facet Srouji, Luis El
On, Mehmet Berkay
Lee, Yun-Jhu
Abdelghany, Mahmoud
Yoo, S. J. Ben
contents Mach-Zehnder Interferometric meshes are attractive for low-loss photonic matrix multiplication but are challenging to program. Using least-squares optimization of directional derivatives, we experimentally demonstrate that desired matrix updates can be implemented agnostic to hardware imperfections. \c{opyright} 2024 The Author(s)
format Preprint
id arxiv_https___arxiv_org_abs_2401_03564
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Experimental Demonstration of Imperfection-Agnostic Local Learning Rules on Photonic Neural Networks with Mach-Zehnder Interferometric Meshes
Srouji, Luis El
On, Mehmet Berkay
Lee, Yun-Jhu
Abdelghany, Mahmoud
Yoo, S. J. Ben
Optics
Systems and Control
Mach-Zehnder Interferometric meshes are attractive for low-loss photonic matrix multiplication but are challenging to program. Using least-squares optimization of directional derivatives, we experimentally demonstrate that desired matrix updates can be implemented agnostic to hardware imperfections. \c{opyright} 2024 The Author(s)
title Experimental Demonstration of Imperfection-Agnostic Local Learning Rules on Photonic Neural Networks with Mach-Zehnder Interferometric Meshes
topic Optics
Systems and Control
url https://arxiv.org/abs/2401.03564