WASP MATH-AI PhD position in probability and analysis
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 the interplay between probability and analysis in the setting of machine learning, under the supervision of Pierre Nyquist.
The subject of the PhD project will be decided together with the student from a range of topics with the overall aim to build towards a mathematical theory for deep learning and learning algorithms. The project sits at the intersection of probability theory, mathematical analysis, specifically partial differential equations and gradient flows, statistical physics and machine learning. The analysis of interacting particle systems will be a central theme, as will the role of stochatic control and stochatic analysis in the setting of learning algorithms.
The position is a time-limited, full-time, fully funded five year position (four years of research + 20% teaching spread out over the duration of the position) starting August 2020 or at an agreed upon time. It comes with generous support for travel and potentially longer stays abroad. The position is financed within the Wallenberg Autonomous Systems and Software Program (WASP) and the student will participate in the WASP graduate school; for more information about the WASP program see https://wasp-sweden.org/graduate-school/ai-graduate-school-courses/.
At KTH, the successful candidate will be part of the group in probability and mathematical statistics, and have strong interactions with the Brummer & Partners MathDataLab, a research lab hosted in the Department of Mathematics that aims at creating a hub for mathematical research in the analysis of complex data.
Students interested in one or more fields related to the following are encouraged to apply: probability theory, artificial intelligence, machine learning, statistics, partial differential equations, gradient flows, interacting particle systems, statistical physics, stochatic analysis. Ideally, applicants should have a sound training in probability theory (including stochastic processes and stochastic differential equations) and analysis (including partial differential equations and functional analysis). Students with deficiencies in one of these areas but who are strong in others are nonetheless urged to apply.
The official application period opens February 13 via KTH's recruitment system, but intersted students are encouraged to contact Pierre Nyquist as soon as possible for a informal conversation or an early application (must include the same material as the official application, see attached PDF); qualified candidates may then be asked to resubmit the same material via KTH Recruitment once the position opens there.
For more information and links to the application system (once open to submissions) see https://people.kth.se/~pierren/open.html.