PhD student position in Machine Learning, formally based at the Division of Statistics and Machine Learning, Linköping University.
Your work assignments
The research focus for the advertised position is machine learning for spatio-temporal modeling, prediction, and reasoning. The project involves the development of novel methods for combining deep-learning-based predictive models (for instance based on graph neural networks, transformers and neural operators) with probabilistic models for representing and reasoning about uncertainties in the models and their predictions (for instance by using latent variables, deep generative models, and ensemble predictions). The project involves both fundamental research on new machine learning models and computational algorithms, as well as applied research to demonstrate the usefulness of the new methods. For the latter, particular emphasis will be put on problems related to weather forecasting and climate science. Examples of this include predicting the risk of extreme weather events, combining predictive models with remotely observed data (e.g. hyperspectral images from satellites), and developing space-time continuous machine learning models. The PhD project will be carried out in collaboration with the Swedish Meteorological and Hydrological Institute (SMHI). Machine learning has made a significant impact on applications related to climate and weather over the past few years and the advertised PhD project will contribute to this development.
The position is part of the project __main__: Multi-dimensional Alignment and Integration of Physical and Virtual Worlds, funded by the Wallenberg AI, Autonomous Systems and Software Program (WASP) through the NEST initiative.