In-situ training in programmable photonic frequency circuits

Optical artificial neural networks (OANNs) leverage the advantages of photonic technologies including high processing speeds, low energy consumption, and mass production to establish a competitive and scalable platform for machine learning applications. While recent advancements have focused on harn...

Full description

Saved in:
Bibliographic Details
Main Authors: Rübeling Philip, Marchukov Oleksandr V., Bellotti Filipe F., Hoff Ulrich B., Zinner Nikolaj T., Kues Michael
Format: Article
Language:English
Published: De Gruyter 2025-06-01
Series:Nanophotonics
Subjects:
Online Access:https://doi.org/10.1515/nanoph-2025-0125
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849234111012536320
author Rübeling Philip
Marchukov Oleksandr V.
Bellotti Filipe F.
Hoff Ulrich B.
Zinner Nikolaj T.
Kues Michael
author_facet Rübeling Philip
Marchukov Oleksandr V.
Bellotti Filipe F.
Hoff Ulrich B.
Zinner Nikolaj T.
Kues Michael
author_sort Rübeling Philip
collection DOAJ
description Optical artificial neural networks (OANNs) leverage the advantages of photonic technologies including high processing speeds, low energy consumption, and mass production to establish a competitive and scalable platform for machine learning applications. While recent advancements have focused on harnessing spatial or temporal modes of light, the frequency domain attracts a lot of attention, with current implementations including spectral multiplexing, neural networks in nonlinear optical systems and extreme learning machines. Here, we present an experimental realization of a programmable photonic frequency circuit, realized with fiber-optical components, and implement the in-situ training with optical weight control of an OANN operating in the frequency domain. Input data is encoded into phases of frequency comb modes, and programmable phase and amplitude manipulations of the spectral modes enable in-situ training of the OANN, without employing a digital model of the device. The trained OANN achieves multiclass classification accuracies exceeding 90 %, comparable to conventional machine learning approaches. This proof-of-concept demonstrates the feasibility of a multilayer OANN in the frequency domain and can be extended to a scalable, integrated photonic platform with ultrafast weights updates, with potential applications to single-shot classification in spectroscopy.
format Article
id doaj-art-552da5922f7a475bafcb4059d6cee709
institution Kabale University
issn 2192-8614
language English
publishDate 2025-06-01
publisher De Gruyter
record_format Article
series Nanophotonics
spelling doaj-art-552da5922f7a475bafcb4059d6cee7092025-08-20T04:03:17ZengDe GruyterNanophotonics2192-86142025-06-0114162779278610.1515/nanoph-2025-0125In-situ training in programmable photonic frequency circuitsRübeling Philip0Marchukov Oleksandr V.1Bellotti Filipe F.2Hoff Ulrich B.3Zinner Nikolaj T.4Kues Michael5Institute of Photonics (IOP), Leibniz University Hannover, Nienburger Str. 17, Hannover, GermanyInstitute of Photonics (IOP), Leibniz University Hannover, Nienburger Str. 17, Hannover, GermanyKvantify Aps, DK-2100Copenhagen, DenmarkKvantify Aps, DK-2100Copenhagen, DenmarkKvantify Aps, DK-2100Copenhagen, DenmarkInstitute of Photonics (IOP), Leibniz University Hannover, Nienburger Str. 17, Hannover, GermanyOptical artificial neural networks (OANNs) leverage the advantages of photonic technologies including high processing speeds, low energy consumption, and mass production to establish a competitive and scalable platform for machine learning applications. While recent advancements have focused on harnessing spatial or temporal modes of light, the frequency domain attracts a lot of attention, with current implementations including spectral multiplexing, neural networks in nonlinear optical systems and extreme learning machines. Here, we present an experimental realization of a programmable photonic frequency circuit, realized with fiber-optical components, and implement the in-situ training with optical weight control of an OANN operating in the frequency domain. Input data is encoded into phases of frequency comb modes, and programmable phase and amplitude manipulations of the spectral modes enable in-situ training of the OANN, without employing a digital model of the device. The trained OANN achieves multiclass classification accuracies exceeding 90 %, comparable to conventional machine learning approaches. This proof-of-concept demonstrates the feasibility of a multilayer OANN in the frequency domain and can be extended to a scalable, integrated photonic platform with ultrafast weights updates, with potential applications to single-shot classification in spectroscopy.https://doi.org/10.1515/nanoph-2025-0125machine learningphotonic computingultrafast optics
spellingShingle Rübeling Philip
Marchukov Oleksandr V.
Bellotti Filipe F.
Hoff Ulrich B.
Zinner Nikolaj T.
Kues Michael
In-situ training in programmable photonic frequency circuits
Nanophotonics
machine learning
photonic computing
ultrafast optics
title In-situ training in programmable photonic frequency circuits
title_full In-situ training in programmable photonic frequency circuits
title_fullStr In-situ training in programmable photonic frequency circuits
title_full_unstemmed In-situ training in programmable photonic frequency circuits
title_short In-situ training in programmable photonic frequency circuits
title_sort in situ training in programmable photonic frequency circuits
topic machine learning
photonic computing
ultrafast optics
url https://doi.org/10.1515/nanoph-2025-0125
work_keys_str_mv AT rubelingphilip insitutraininginprogrammablephotonicfrequencycircuits
AT marchukovoleksandrv insitutraininginprogrammablephotonicfrequencycircuits
AT bellottifilipef insitutraininginprogrammablephotonicfrequencycircuits
AT hoffulrichb insitutraininginprogrammablephotonicfrequencycircuits
AT zinnernikolajt insitutraininginprogrammablephotonicfrequencycircuits
AT kuesmichael insitutraininginprogrammablephotonicfrequencycircuits