News

Mathematician receives prestigious grant in machine learning

  • Faculté des Sciences, des Technologies et de Médecine (FSTM)
    Université / Administration centrale et Rectorat
    30 août 2021
  • Catégorie
    Recherche, Université
  • Thème
    Mathématiques

Mikołaj Kasprzak, postdoctoral researcher within the Department of Mathematics at the University of Luxembourg has been recently selected for the Marie Skłodowska-Curie Global Fellowship programme. During two years, this fellowship will enable Mikolaj to investigate machine learning in three institutions, the Massachusetts Institute of Technology, the University College London and the University of Luxembourg.

Mikołaj Kasprzak is working within the research group Probability Theory and Mathematical Finance, led by Prof. Giovanni Peccati, Head of the Department of Mathematics at the University of Luxembourg.

“The Marie Skłodowska-Curie Global Fellowship obtained by Mikolaj is also a very prestigious achievement for the department. Mikolaj has been working in my group for the last couple of years, where he has performed stellar research in several areas of modern probability. It has been an honour to collaborate with him on a number of projects. His exciting success story will serve as further motivation for us to continue hiring top young collaborators from around the globe, and to increase more and more our ability to attract competitive European funding”, comments Prof. Peccati.

Mikołaj Kasprzak explains his project in more detail.

What is your research about?

“The aim of my research is to advance the development of quality measures for approximations in machine learning and statistics, using the rich theoretical machinery of mathematical analysis and probability. Approximate inference has developed significantly in the recent years as a way to speed up the learning process of algorithms based on statistics. This is particularly important in the era of big data, in which the size of datasets used in applications often makes it difficult to draw accurate conclusions in a reasonable amount of time. Researchers and practitioners therefore look for tools to measure both the efficiency of the associated inference and sampling methods and the size of the error they generate when using approximations. Indeed, such tools are still not widely available in the literature for many of the commonly used models. At the same time, probability theory and functional analysis provide the machinery that could lead to a real breakthrough in this area, if applied to the most important problems in a thoughtful manner. I hope that my expertise will allow me to do exactly that and thus contribute to the development of this field.”

What will you do during two years?

“During the two years I will work on a variety of projects in a collaboration with researchers from the three host institutions involved in the project. I will spend my time at MIT and UCL concentrating mostly on the applied aspects of the project, using in practice the theory that I have become expert in during my PhD and postdoc. While working at the University of Luxembourg I plan to develop new results in probability and mathematical statistics that will be inspired by applications to machine learning. Overall, however, I hope to combine working on both the theoretical and applied aspects of my project throughout the duration of fellowship and write joint papers with collaborators from all the three institutions, thus bringing together the strengths of the three research groups I will be part of. Moreover, I plan to participate in various outreach activities and get involved in graduate teaching at the University of Luxembourg.”

What does this award mean to you?

“The fellowship will let me broaden my research agenda, experience working in different countries and with different teams and make my research truly multidisciplinary. It will help me become a more mature academic and will certainly increase my chances of securing a good permanent position in the future. The fellowship will be a great learning experience that will significantly accelerate my career development. It will also help establish a long-term collaboration between the institutions involved in the project that will lead to many new exciting research developments.”