We review our recent progress on the application of machine-learning techniques in the field of ultrafast nonlinear fibre optics. We demonstrate that neural networks can both efficiently predict the temporal and spectral features of optical signals that are obtained after propagation in the presence of focusing and defocusing nonlinearity and solve the associated inverse problem. We also show that evolutionary algorithms can be used to control complex nonlinear dynamics of ultrafast fibre lasers.
|Title of host publication||Proceedings of 20th International Conference Laser Optics (ICLO 2022)|
|Number of pages||1|
|Publication status||Published - Jun 2022|
|Event||20th International Conference Laser Optics - Saint Petersburg, Russian Federation|
Duration: 20 Jun 2022 → 24 Jun 2022
|Conference||20th International Conference Laser Optics|
|Period||20/06/22 → 24/06/22|