Mini-Symposia: ‘Introduction to Machine Learning’
Two weeks ago (7th-8th September 2020), I attended an online mini-symposia that was co-organized by four different H2020 ITN projects (FONTE, REAL-NET, MOCCA, and WON) about machine learning in optical communications.
It was a 2-day symposium where the students could enjoy on the first day am enlightening introduction about machine learning from Prof. Dr. Darko Zibar, with hands-on in small problems and discussions afterward. Right next, Dr. Kamalian-Kopae Morteza showed us different Machine learning-based equalization in fiber-optic communication, and to close the first day, Dr. Milad Sefidgaran showed us the information theory of the optical fiber. We were exposed to several equations and discussions, which was marvelous to me.
The second day Prof. Dr. David Saad gave us a remarkable didact talk about different approaches to machine learning techniques (Bayesian approach, Support vector machines, Markovian dynamics, and much more). Dr. Jelena Pesic & Dr. Matteo Lonardi from Nokia Bell Labs France talked to us about the trade-of of using machine learning in optical communications. Specifically, answer the following question: Will Machine Learning mitigate the extra cost of an increase in capacity in optical fiber?
Finishing the symposium, Prof. Dr. Nathan Kutz (University of Washington) presented us with an exceptional technique to find an optimal coordinate system and to discover governing equations, especially for dynamic systems.
I am grateful to have had the opportunity to learn so much and to get in contact with so many competent professionals in our field. Hope to keep in touch!
