Engineering Spectro-Temporal Light States with Physics-Embedded Deep Learning

Frequency synthesis and spectro-temporal control of optical wavepackets are central to ultrafast science, with supercontinuum (SC) generation standing as one remarkable example. Through passive manipulation, femtosecond pulses from nanojoule-level lasers can be transformed into octave-spanning spect...

Full description

Saved in:
Bibliographic Details
Main Authors: Shilong Liu, Stéphane Virally, Gabriel Demontigny, Patrick Cusson, Denis V. Seletskiy
Format: Article
Language:English
Published: American Association for the Advancement of Science (AAAS) 2025-01-01
Series:Ultrafast Science
Online Access:https://spj.science.org/doi/10.34133/ultrafastscience.0107
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849239354466107392
author Shilong Liu
Stéphane Virally
Gabriel Demontigny
Patrick Cusson
Denis V. Seletskiy
author_facet Shilong Liu
Stéphane Virally
Gabriel Demontigny
Patrick Cusson
Denis V. Seletskiy
author_sort Shilong Liu
collection DOAJ
description Frequency synthesis and spectro-temporal control of optical wavepackets are central to ultrafast science, with supercontinuum (SC) generation standing as one remarkable example. Through passive manipulation, femtosecond pulses from nanojoule-level lasers can be transformed into octave-spanning spectra, supporting few-cycle pulse outputs when coupled with external pulse compressors. While strategies such as machine learning have been applied to control the SC’s central wavelength and bandwidth, their success has been limited by the nonlinearities and strong sensitivity to measurement noise. Here, we propose and demonstrate how a physics-embedded convolutional neural network that embeds spectro-temporal correlations can circumvent such challenges, resulting in faster convergence and reduced noise sensitivity. This innovative approach enables on-demand control over spectro-temporal features of SC, achieving few-cycle pulse shaping without external compressors. This approach heralds a new era of arbitrary spectro-temporal light state engineering, with implications for ultrafast photonics, photonic neuromorphic computation, and artificial intelligence-driven optical systems.
format Article
id doaj-art-19748e938bb94328a55fedb85daa1dc7
institution Kabale University
issn 2765-8791
language English
publishDate 2025-01-01
publisher American Association for the Advancement of Science (AAAS)
record_format Article
series Ultrafast Science
spelling doaj-art-19748e938bb94328a55fedb85daa1dc72025-08-20T04:01:02ZengAmerican Association for the Advancement of Science (AAAS)Ultrafast Science2765-87912025-01-01510.34133/ultrafastscience.0107Engineering Spectro-Temporal Light States with Physics-Embedded Deep LearningShilong Liu0Stéphane Virally1Gabriel Demontigny2Patrick Cusson3Denis V. Seletskiy4femtoQ Lab, Department of Engineering Physics, Polytechnique Montréal, Montréal, Québec H3T 1J4, Canada.femtoQ Lab, Department of Engineering Physics, Polytechnique Montréal, Montréal, Québec H3T 1J4, Canada.femtoQ Lab, Department of Engineering Physics, Polytechnique Montréal, Montréal, Québec H3T 1J4, Canada.femtoQ Lab, Department of Engineering Physics, Polytechnique Montréal, Montréal, Québec H3T 1J4, Canada.femtoQ Lab, Department of Engineering Physics, Polytechnique Montréal, Montréal, Québec H3T 1J4, Canada.Frequency synthesis and spectro-temporal control of optical wavepackets are central to ultrafast science, with supercontinuum (SC) generation standing as one remarkable example. Through passive manipulation, femtosecond pulses from nanojoule-level lasers can be transformed into octave-spanning spectra, supporting few-cycle pulse outputs when coupled with external pulse compressors. While strategies such as machine learning have been applied to control the SC’s central wavelength and bandwidth, their success has been limited by the nonlinearities and strong sensitivity to measurement noise. Here, we propose and demonstrate how a physics-embedded convolutional neural network that embeds spectro-temporal correlations can circumvent such challenges, resulting in faster convergence and reduced noise sensitivity. This innovative approach enables on-demand control over spectro-temporal features of SC, achieving few-cycle pulse shaping without external compressors. This approach heralds a new era of arbitrary spectro-temporal light state engineering, with implications for ultrafast photonics, photonic neuromorphic computation, and artificial intelligence-driven optical systems.https://spj.science.org/doi/10.34133/ultrafastscience.0107
spellingShingle Shilong Liu
Stéphane Virally
Gabriel Demontigny
Patrick Cusson
Denis V. Seletskiy
Engineering Spectro-Temporal Light States with Physics-Embedded Deep Learning
Ultrafast Science
title Engineering Spectro-Temporal Light States with Physics-Embedded Deep Learning
title_full Engineering Spectro-Temporal Light States with Physics-Embedded Deep Learning
title_fullStr Engineering Spectro-Temporal Light States with Physics-Embedded Deep Learning
title_full_unstemmed Engineering Spectro-Temporal Light States with Physics-Embedded Deep Learning
title_short Engineering Spectro-Temporal Light States with Physics-Embedded Deep Learning
title_sort engineering spectro temporal light states with physics embedded deep learning
url https://spj.science.org/doi/10.34133/ultrafastscience.0107
work_keys_str_mv AT shilongliu engineeringspectrotemporallightstateswithphysicsembeddeddeeplearning
AT stephanevirally engineeringspectrotemporallightstateswithphysicsembeddeddeeplearning
AT gabrieldemontigny engineeringspectrotemporallightstateswithphysicsembeddeddeeplearning
AT patrickcusson engineeringspectrotemporallightstateswithphysicsembeddeddeeplearning
AT denisvseletskiy engineeringspectrotemporallightstateswithphysicsembeddeddeeplearning