The Department of Mathematics at KTH Royal Institute of Technology in Stockholm has an opening for a PhD position in applied and computational mathematics, with a focus on high dimensional statistical learning for AI, under the supervision of Tatjana Pavlenko.

The overall goal of this project is to build towards a mathematical theory and novel computational methodologies of advanced and efficient high-dimensional statistical inference that are able to accommodate structured sparsity of the data. A specific aspect of the Big Data setting we see today is that many of automatically measured features have little relevance to a given project. This poses a needlein-a-haystack problem: a very few valuable features must be detected among huge amount of useless. The combination of such type of structured sparsity with high-dimensionality is a key challenge of the modern AI and machine learning applications.

The methods of high-dimensional probability have been crucial for recent progress in understanding the performance of new statistical learning algorithms and for designing of contemporary methods of analysis of big data.

The successful candidate will pursue a PhD project at the intersection of the probability theory and statistics in high and infinite dimensions, sparse statistical modeling, empirical processes theory, random matrix theory and artificial intelligence. Leveraging the interplay between the theory of adaptive signal detection in sparse and weak settings, structured sparsity learning, infinite-dimensional asymptotic theory, weighted empirical processes and topics within the random matrix theory such as low rank matrix recovery and other types of structural patterns, will be a central subject of the project. Once the mathematical foundations have been laid, the focus will be placed on devising and implementing fully adaptive, data-driven statistical and algorithmic inferential procedures.

Students interested in one or more fields related to the following are encouraged to apply: high dimensional statistical inference, empirical processes, theory of statistical learning with sparsity, artificial intelligence, graphical modeling and random matrix theory.

The position is a time-limited, full-time, five year position starting August 2020 or at an agreed upon date. The position is fully funded for four years and will be extended to five years by assigning teaching duties. It also includes generous travel support. The position is funded within the Wallenberg Autonomous Systems and Software Program (WASP), and the student will participate in the WASP graduate school. Through this program the student will have a wide variety of opportunities to interact with other researchers and industry collaborators in AI, ML, and statistics, including events such as conferences and PhD courses; see

https://wasp-sweden.org/graduate-school/ai-graduate-school-courses/

for more information about the WASP program.

The successful candidate will be part of two divisions within the Department of Mathematics, Mathematical Statistics and Mathematics of Data and AI.

For more information and links to the application system see: https://people.kth.se/~pavlenko/