An article by NSU academics was among the most downloaded articles in the highly rated (impact factor 3.866) Optics Letters scientific journal in July 2019. The article described the results of their research on a new method for determining the duration of ultrashort laser pulses based on machine learning. The article reference is: “Kokhanovskiy A., Bednyakova A., Kuprikov E., Ivanenko A., Dyatlov M., Lotkov D., Kobtsev S., Turitsyn S. Machine learning-based pulse characterization in figure-eight mode-locked lasers // Opt . Lett. 2019. Vol. 44. Iss. 13. P. 3410–3413
Alexei Kokhanovsky, an NSU graduate student and one of the authors, talked about their methodology,
The special instruments that are typically used to determine the duration of ultrashort laser pulses are expensive and not easy to use. We figured out how to measure the duration of picosecond pulses with a nanosecond time resolution photodetector and a combination of machine learning methods with the dispersion Fourier transform.
A unique experimental facility was created at the NSU Division of Laser Physics and Innovative Technologies that confirmed the advantages of this new method.
Sergey Kobtsev, Head of the NSU Physics Department’s Division of Laser Physics and Innovative Technologies, described why this research is of interest,
Fiber lasers that generate ultrashort radiation pulses are a new type of laser that are currently being studied all over the world. The new scientific results from studying the generation regimes of these lasers and determining the parameters of their radiation are of great interest to laser experts as well as scientists and developers from other spheres that use lasers in their activities.
This work was conducted based on a collaboration between Novosibirsk State University, the Institute of Computational Technologies SB RAS, and the Aston Institute of Photonics Technology (Great Britain).
Professor Sergey Turitsyn, Director of the Nonlinear Photonics Laboratory at the NSU Department of Physics and Director of the Aston Institute of Photonics Technology, provided more details,
Machine learning methods, which are often mentioned in the much wider context of artificial intelligence research, are expanding very quickly into almost all areas of modern science. The use of machine learning algorithms in photonics allows one to reach a fundamentally new level of advanced laser systems design that generate radiation with the required properties. NSU is one of the world leaders in this emerging new field of laser physics.
