Deep learning assisted engineering of supercontinuum and few-cycle laser pulses
We demonstrate a machine-learning approach to controlled supercontinuum and few-cycle laser engineering with nJ-level inputs. Assisted by a physics-trained convolutional neural network, we achieve these spectrotemporal manipulations: three-cycle pulses, controlled dispersive wave and high-order solitons.
Authors
- Denis V. Seletskiy (ORCID: https://orcid.org/0000-0003-3480-4595)
- Shilong Liu (ORCID: https://orcid.org/0000-0003-4582-783X)
- Stéphane Virally (ORCID: https://orcid.org/0000-0002-8200-1714)
- Patrick Cusson
- Gabriel Demontigny
Publication Details
- Published
- 2025-01-01
- DOI
- https://doi.org/10.1364/cleo_at.2025.jps200_32
- Primary Topic
- Laser-Matter Interactions and Applications
- Type
- article
- Field-Weighted Citation Impact
- 0.00