A Deep Reinforcement Learning Algorithm for Smart Control of Hysteresis Phenomena in a Mode-Locked Fiber Laser

Alexey Kokhanovskiy; Alexey Shevelev; Kirill Serebrennikov; Evgeny Kuprikov; Sergei K. Turitsyn

doi:10.3390/photonics9120921

A Deep Reinforcement Learning Algorithm for Smart Control of Hysteresis Phenomena in a Mode-Locked Fiber Laser

Alexey Kokhanovskiy^*, Alexey Shevelev, Kirill Serebrennikov, Evgeny Kuprikov, Sergei K. Turitsyn

^*Corresponding author for this work

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

Abstract

We experimentally demonstrate the application of a double deep Q-learning network algorithm (DDQN) for design of a self-starting fiber mode-locked laser. In contrast to the static optimization of a system design, the DDQN reinforcement algorithm is capable of learning the strategy of dynamic adjustment of the cavity parameters. Here, we apply the DDQN algorithm for stable soliton generation in a fiber laser cavity exploiting a nonlinear polarization evolution mechanism. The algorithm learns the hysteresis phenomena that manifest themselves as different pumping-power thresholds for mode-locked regimes for diverse trajectories of adjusting optical pumping.

Original language	English
Article number	921
Number of pages	7
Journal	Photonics
Volume	9
Issue number	12
Early online date	30 Nov 2022
DOIs	https://doi.org/10.3390/photonics9120921
Publication status	Published - Dec 2022

Bibliographical note

Copyright © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and
conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

Data Access Statement

Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.

Keywords

fibre mode-locked lasers
reinforcement learning
hysteresis phenomena

Access to Document

10.3390/photonics9120921Licence: CC BY 4.0

A Kokhanovskiy et al_photonics-09-00921_Deep Reinforcement Learning Algorithm for Smart Control of Hysteresis Phenomena in a Mode-Locked Fiber Laser
Copyright © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Final published version, 9.22 MBLicence: CC BY 4.0

Cite this

@article{85022c2d0e6e42c0964098b3343e6911,

title = "A Deep Reinforcement Learning Algorithm for Smart Control of Hysteresis Phenomena in a Mode-Locked Fiber Laser",

abstract = "We experimentally demonstrate the application of a double deep Q-learning network algorithm (DDQN) for design of a self-starting fiber mode-locked laser. In contrast to the static optimization of a system design, the DDQN reinforcement algorithm is capable of learning the strategy of dynamic adjustment of the cavity parameters. Here, we apply the DDQN algorithm for stable soliton generation in a fiber laser cavity exploiting a nonlinear polarization evolution mechanism. The algorithm learns the hysteresis phenomena that manifest themselves as different pumping-power thresholds for mode-locked regimes for diverse trajectories of adjusting optical pumping.",

keywords = "fibre mode-locked lasers, reinforcement learning, hysteresis phenomena",

author = "Alexey Kokhanovskiy and Alexey Shevelev and Kirill Serebrennikov and Evgeny Kuprikov and Turitsyn, {Sergei K.}",

note = "Copyright {\textcopyright} 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).",

year = "2022",

month = dec,

doi = "10.3390/photonics9120921",

language = "English",

volume = "9",

journal = "Photonics",

issn = "2304-6732",

publisher = "MDPI AG",

number = "12",

}

TY - JOUR

T1 - A Deep Reinforcement Learning Algorithm for Smart Control of Hysteresis Phenomena in a Mode-Locked Fiber Laser

AU - Kokhanovskiy, Alexey

AU - Shevelev, Alexey

AU - Serebrennikov, Kirill

AU - Kuprikov, Evgeny

AU - Turitsyn, Sergei K.

N1 - Copyright © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).

PY - 2022/12

Y1 - 2022/12

N2 - We experimentally demonstrate the application of a double deep Q-learning network algorithm (DDQN) for design of a self-starting fiber mode-locked laser. In contrast to the static optimization of a system design, the DDQN reinforcement algorithm is capable of learning the strategy of dynamic adjustment of the cavity parameters. Here, we apply the DDQN algorithm for stable soliton generation in a fiber laser cavity exploiting a nonlinear polarization evolution mechanism. The algorithm learns the hysteresis phenomena that manifest themselves as different pumping-power thresholds for mode-locked regimes for diverse trajectories of adjusting optical pumping.

AB - We experimentally demonstrate the application of a double deep Q-learning network algorithm (DDQN) for design of a self-starting fiber mode-locked laser. In contrast to the static optimization of a system design, the DDQN reinforcement algorithm is capable of learning the strategy of dynamic adjustment of the cavity parameters. Here, we apply the DDQN algorithm for stable soliton generation in a fiber laser cavity exploiting a nonlinear polarization evolution mechanism. The algorithm learns the hysteresis phenomena that manifest themselves as different pumping-power thresholds for mode-locked regimes for diverse trajectories of adjusting optical pumping.

KW - fibre mode-locked lasers

KW - reinforcement learning

KW - hysteresis phenomena

UR - https://www.mdpi.com/2304-6732/9/12/921

U2 - 10.3390/photonics9120921

DO - 10.3390/photonics9120921

M3 - Article

SN - 2304-6732

VL - 9

JO - Photonics

JF - Photonics

IS - 12

M1 - 921

ER -

A Deep Reinforcement Learning Algorithm for Smart Control of Hysteresis Phenomena in a Mode-Locked Fiber Laser

Abstract

Bibliographical note

Data Access Statement

Keywords

Access to Document

Other files and links

Fingerprint

Cite this