Deep learning reconstruction of ultrashort pulses

Tom Zahavy; Alex Dikopoltsev; Daniel Moss; Gil Ilan Haham; Oren Cohen; Shie Mannor; Mordechai Segev

doi:10.1364/OPTICA.5.000666

Deep learning reconstruction of ultrashort pulses

Tom Zahavy, Alex Dikopoltsev^*, Daniel Moss, Gil Ilan Haham, Oren Cohen, Shie Mannor, Mordechai Segev

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

132 Scopus citations

Abstract

Ultrashort laser pulses with femtosecond to attosecond pulse duration are the shortest systematic events humans can currently create. Characterization (amplitude and phase) of these pulses is a crucial ingredient in ultrafast science, e.g., exploring chemical reactions and electronic phase transitions. Here, we propose and demonstrate, numerically and experimentally, what is to the best of our knowledge, the first deep neural network technique to reconstruct ultrashort optical pulses. Employing deep neural networks for reconstruction of ultrashort pulses enables diagnostics of very weak pulses and offers new possibilities, e.g., reconstruction of pulses using measurement devices without knowing in advance the relations between the pulses and the measured signals. Finally, we demonstrate the ability to reconstruct ultrashort pulses from their experimentally measured frequency-resolved optical gating traces via deep networks that have been trained on simulated data.

Original language	English
Pages (from-to)	666-673
Number of pages	8
Journal	Optica
Volume	5
Issue number	5
DOIs	https://doi.org/10.1364/OPTICA.5.000666
State	Published - 20 May 2018
Externally published	Yes

Access to Document

10.1364/OPTICA.5.000666

Cite this

@article{ac12a6b9cbf841e4bbbbb5efc478ed6e,

title = "Deep learning reconstruction of ultrashort pulses",

abstract = "Ultrashort laser pulses with femtosecond to attosecond pulse duration are the shortest systematic events humans can currently create. Characterization (amplitude and phase) of these pulses is a crucial ingredient in ultrafast science, e.g., exploring chemical reactions and electronic phase transitions. Here, we propose and demonstrate, numerically and experimentally, what is to the best of our knowledge, the first deep neural network technique to reconstruct ultrashort optical pulses. Employing deep neural networks for reconstruction of ultrashort pulses enables diagnostics of very weak pulses and offers new possibilities, e.g., reconstruction of pulses using measurement devices without knowing in advance the relations between the pulses and the measured signals. Finally, we demonstrate the ability to reconstruct ultrashort pulses from their experimentally measured frequency-resolved optical gating traces via deep networks that have been trained on simulated data.",

author = "Tom Zahavy and Alex Dikopoltsev and Daniel Moss and Haham, {Gil Ilan} and Oren Cohen and Shie Mannor and Mordechai Segev",

note = "Publisher Copyright: {\textcopyright} 2018 Optical Society of America.",

year = "2018",

month = may,

day = "20",

doi = "10.1364/OPTICA.5.000666",

language = "英语",

volume = "5",

pages = "666--673",

journal = "Optica",

issn = "2334-2536",

publisher = "OSA Publishing",

number = "5",

}

TY - JOUR

T1 - Deep learning reconstruction of ultrashort pulses

AU - Zahavy, Tom

AU - Dikopoltsev, Alex

AU - Moss, Daniel

AU - Haham, Gil Ilan

AU - Cohen, Oren

AU - Mannor, Shie

AU - Segev, Mordechai

PY - 2018/5/20

Y1 - 2018/5/20

N2 - Ultrashort laser pulses with femtosecond to attosecond pulse duration are the shortest systematic events humans can currently create. Characterization (amplitude and phase) of these pulses is a crucial ingredient in ultrafast science, e.g., exploring chemical reactions and electronic phase transitions. Here, we propose and demonstrate, numerically and experimentally, what is to the best of our knowledge, the first deep neural network technique to reconstruct ultrashort optical pulses. Employing deep neural networks for reconstruction of ultrashort pulses enables diagnostics of very weak pulses and offers new possibilities, e.g., reconstruction of pulses using measurement devices without knowing in advance the relations between the pulses and the measured signals. Finally, we demonstrate the ability to reconstruct ultrashort pulses from their experimentally measured frequency-resolved optical gating traces via deep networks that have been trained on simulated data.

AB - Ultrashort laser pulses with femtosecond to attosecond pulse duration are the shortest systematic events humans can currently create. Characterization (amplitude and phase) of these pulses is a crucial ingredient in ultrafast science, e.g., exploring chemical reactions and electronic phase transitions. Here, we propose and demonstrate, numerically and experimentally, what is to the best of our knowledge, the first deep neural network technique to reconstruct ultrashort optical pulses. Employing deep neural networks for reconstruction of ultrashort pulses enables diagnostics of very weak pulses and offers new possibilities, e.g., reconstruction of pulses using measurement devices without knowing in advance the relations between the pulses and the measured signals. Finally, we demonstrate the ability to reconstruct ultrashort pulses from their experimentally measured frequency-resolved optical gating traces via deep networks that have been trained on simulated data.

UR - http://www.scopus.com/inward/record.url?scp=85047245903&partnerID=8YFLogxK

U2 - 10.1364/OPTICA.5.000666

DO - 10.1364/OPTICA.5.000666

M3 - 文章

AN - SCOPUS:85047245903

SN - 2334-2536

VL - 5

SP - 666

EP - 673

JO - Optica

JF - Optica

IS - 5

ER -

Deep learning reconstruction of ultrashort pulses

Abstract

Access to Document

Other files and links

Fingerprint

Cite this