PhD students in interdisciplinary mathematics

Current PhD students

Ali Mesbah

Department of Electrical Engineering
Admitted: 2026
Project description: The project investigates the development of learning-based, distributed control methods for multi-robot systems operating in highly uncertain and dynamic environments. The focus is on coordinated on-the-fly learning of unknown multi-agent dynamics, safety-critical control with formal guarantees, and optimal performance, particularly under faults and limited communication. The approach shall combine set-theoretic learning, barrier function-based safety, and distributed control algorithms, exploiting basic physical principles and cooperation between healthy and faulty agents.

Pritish Ranjan Joshi

Department: Information Technology
Admitted: 2025
Project description: My research project, Inference-Guided Multiscale Modeling of Metal Plating Dynamics, focuses on expediting the study of stable and effective metal deposition through multiphysics simulations involving coupled partial differential equations (PDEs). The goal is to solve complex numerical systems representing mechanical, morphological, and electrochemical interactions by developing machine learning methods and to utilize simulation-based inference for predictive insights. This work has the potential to contribute to the development of high-energy-density, sustainable rechargeable batteries.

Rafael Rodriguez Velasco

Department: Information Technology
Admitted: 2025
Project description: My research project, Optimal Control of Fusion Processes, focuses on highly accurate magnetohydrodynamic (MHD) simulations using finite element methods. The goal is to solve optimization problems for plasma stabilization and to explore deep reinforcement learning for real-time control. This work has the potential to contribute to the development of feasible fusion energy reactors: a potentially limitless source of sustainable energy

Dhanushki Mapitagama

Department: Information Technology
Admitted: 2024
Project description: The project aims to develop adaptive design of large-scale scientific experiments, focusing on phenotypic drug combination screening for drug repurposing. The combinatorial nature of the screening problem makes exhaustive searches infeasible. We aim to accelerate the identification of promising compounds by integrating probabilistic deep learning with active learning and sequential sampling to predict potentially synergistic combinations. This deep active learning approach therefore aims to reduce both the cost and time required to discover potential cures for targeted conditions.

Isak Sundelius

Department: Mathematics
Admitted: 2024
Project description: The project aims at utilising tools from algebraic geometry to better understand G_2-manifolds, a type of manifold used in physics in string and M-theory. Put simply, a G_2-manifold is a manifold whose symmetries are encoded in a group of the exceptional type G_2, and such a manifold admits what is called a G_2-structure. In particular, the project aims at better understanding and classifying these G_2-structures.

Merlijn van Emmerik

Department: Mathematics
Admitted: 2024
Project description: My project is concerned with understanding quantum field theories (QFTs), using generalized categorical symmetries, extending the notion of symmetries acting on local operators, to that of symmetries which also act on extended objects (lines, surfaces, etc). In QFT, this implies that every symmetry is associated to a topological defect. The connection between topology and symmetries leads to the framework of topological quantum field theory. These both provide insides in QFT.

Henri Doerks

Department: Mathematics
Admitted: 2024
Project description: The purpose of my current project is to use stochastic control and optimal stopping methods, along with applied and experimental game theory, to study various problems in economics and operations research. The unifying theme of the considered problems is incomplete information where the learning rate of the unknown information is controlled. Thereby, a trade-off between learning about the information and earning is created (or exploration vs exploitation).

Moritz Walden

Department: Physics and Astronomy
Admitted: 2023
Project description: The project is aimed towards using machine learning as a tool in string theory and mathematics. For the physics part, the goal is to construct different string compactifications. There are also fundamental open questions about the geometry underlying these constructions. Hence, the mathematical part of the project aims at charting the structures arising from these questions. Conversely, these string compactifications will be used as a testing ground for machine learning.

Roman Denkin

Department: Information Technology
Admitted: 2023
Project description: The project is dedicated to improving performance and therapeutic value of an ultrasound imaging, by applying machine learning to all stages of ultrasound image forming. An attempt will be made to quantify various tissue properties using ultrasound, like local speed of sound, scattering and attenuation, which may provide a new angle for medical professionals to diagnose and evaluate multiple diseases, like breast cancer or NAFLD (non-alcoholic fatty liver disease).

Harald Agelii

Department: Physics and Astronomy
Admitted: 2023
Project description: Mass spectrometry and X-ray imaging of single gas-phase bio-molecules provide valuable insight into both molecular structure and function. The possibility of controlling the orientation of the molecule during imaging would improve such measuring techniques. This project aims to investigate the feasibility of orienting polarisable bio-molecules using external electric fields. This includes developing a mathematical model for simulation of the orientation process on time scales beyond the capabilities of existing methods.

Ask Ellingsen

Department: Mathematics
Admitted: 2022
Project description: Broadly, I do mathematical quantum physics. Specifically, my PhD project focuses on an exotic type of particle behaviour called fractional quantum statistics, and on anyons, the particles that display this behaviour. Anyons can be thought of as lying 'between' bosons and fermions, and only appear in certain two-dimensional systems. We are interested in the fundamental mathematical description of anyons, as well as predicting physical properties; especially of the anyon gas.

Yoann Sohnle

Department: Physics and Astronomy
Admitted: 2022

Hannes Gustafsson

Department: Chemistry - Ångström
Admitted: 2022

David Meadon

Department: IT
Admitted: 2022
Project description: My PhD project involves the spectral analysis of block Toeplitz and Toeplitz-like matrices. More specifically, to investigate using the theory of generalised locally Toeplitz (GLT) sequences and matrix-less methods to study the eigenvalues and eigenvectors of different Toeplitz(-like) matrices. For example, these types of matrices arise in the discretisation of differential equations and understanding their spectral properties is important for analysing existing and designing new numerical methods and solvers.

Gesina Menz

Dept: IT
Admitted: 2022
Project description: My PhD project is aiming at modelling living cells in a data-driven way. Since this is a very broad and intensive task, two specific projects I am working on right now include investigating how cell signalling works mechanistically as well as exploring how blood vessel formation can be modelled. Long-term, the focus is mechanistic modelling of different aspects of cell behaviour on a population level.

Andreas Michael

Dept: IT
Admitted: 2021
Project description: My PhD project is part of the larger research project INVIVE which aims to create real-time simulations of the respiratory function of an ICU patient. In my PhD I focus on the simulation of the human diaphragm. This includes creating an accurate model for muscle tissue which accounts for muscle fibre orientation and activation dynamics. Additionally, it includes numerically solving the model equations using methods based on radial basis functions.

Sanya Karilanova

Dept: Electrical Engineering
Admitted: 2021
Project description: Signal Processing with Spiking Neural Networks (SNN). One aspect of this project is to research the development of efficient training and learning method using SNN as that is the current challenge regarding SNN. Furthermore, applying the developed SNN to a data produced by electronic skin.

Li Ju

Dept: IT
Admitted: 2021
Project description: My PhD project focuses on federated machine learning and distributed machine learning. The aim of the project is to develop distributed optimization algorithms with better convergence for federated training, and to enhance the security of federated learning with differential privacy. With algorithmic insights, the project also aims to improve organ segmentation and tumor segmentation for radiation treatment planning with federated learning.

Swarnadip Chatterjee

Dept: IT
Admitted: 2021
Project description: My PhD project is about improving the understanding of the disease and its progression, as well as enabling reliable early detection of cancer. The project will explore and develop techniques for multimodal information fusion, utilizing combinations of several imaging modalities to maximize information gain.The project combines the power of modern deep learning techniques with novel distance measures between images, to create new methods for efficient fusion of multimodal image data.

Alfred Andersson

Dept: Cell and Molecular Biology
Admitted: 2020
Project description: To predict chemical properties, it is crucial to find accurate energy potential functions that describe the interactions between the atoms involved. In reality we do not know what the best representation of these would be and few improvements have been made since the first models were introduced 50 years ago. Today we have access to large datasets and we will use these to find an even more accurate representation.