Mission
We study and advance the foundations of data-driven science and engineering.
We power scientific discovery, driven by AI and statistical science.
We are a group hosted at the Division of Scientific Computing, Science for Life Laboratory (SciLifeLab).
We work on the fundamentals of machine learning and AI, with a strong focus on solving hard problems in (life) sciences.
Funding agencies enabling our research:
Recent News
– Feb 15, 2024: We welcome applications for a PhD position in Scientific Machine Learning (keywords: variational inference, large-scale optimization, robust learning, Bayesian inference). Apply here by April 2, 2024!
– Feb 15, 2024: We welcome applications for a PhD position in Deep Learning for Drug Repurposing (keywords: Bayesian neural networks, large-scale optimization, active learning, drug discovery). Apply here by March 22, 2024! Please select Project 11 titled ‘Adaptive Deep Learning of Drug Combination Mechanics for Accelerated Repurposing‘.
– Feb 13, 2024: Prashant will deliver his Docentship lecture on ‘Global Optimisation of Computationally Expensive Objective Functions’ on 27/02 at 13:00 in Theatrum Visuale (Room 100155), Ångströmlaboratoriet
– Dec 01, 2023: Preprint out! Our take on parameter-free robust learning via variational inference – arXiv link.
– Nov 07, 2023: Swedish Research Council (VR) starting grant awarded to Prashant Singh.
– Oct 23, 2023: We welcome Andrey Shternshis as a PostDoc in our lab.
Spotlight: Jan 2024
Recent student projects:
– Bayesian Sequential Model Optimization for Drug Combination Repurposing (Dhanushki Mapitigama, Mina Badri, Ema Duljkovic)
– Bayesian Optimization for Characterising Quantum Entanglement (Stefanos Tsampanakis, Ramin Modaresi, Niklas Kostrzewa)
– Deep Learning for Ill-Posed Inverse Problems in Photonics (Johan Rensfeldt, Fredrik Gillgren, Gustav Fredrikson)
Research Areas
Sampling
statistical sampling, data gathering, active learning, design of experiments, sequential design, reinforcement learning
Modeling
surrogate modeling, multi-fidelity modeling, deep learning, Bayesian models, time series modeling
Optimization
Bayesian optimization, non-convex optimization, variational inference, surrogate-based single and multi-objective optimization
Inverse Problems
likelihood-free parameter estimation, approximate Bayesian computation, learning summary statistics, ill-posed problems