Uppsala University and the Division of Systems and Control is seeking a Postdoc in Machine Learning with a focus on models evolving over space and time.
The successful candidate will form a part of the NEST initiative funded by the Wallenberg AI, Autonomous Systems and Software Program (WASP) and the Wallenberg National Program for Data-Driven Life Science (DDLS). The project, Learning 3D Genome Dynamics from Heterogeneous Data, is a 5-year collaboration between researchers at Uppsala University and Karolinska Institute.
Project description
This project focuses on developing, analyzing and using probabilistic methods for dynamic phenomena evolving over space and time based on measurements from different and complementary sources. We will develop generally applicable machine learning models and methods driven by the data-rich experiments from our collaborators. The real-world use-case is to learn the rules that govern the dynamics of the bacterial chromosome structure.
Technical building blocks could include state-space models, generative models in the form of diffusion models, deep learning, optimal transport, and probabilistic modelling in general. Computer vision can also be included if there is interest.
The postdoc position involves the development of theory and probabilistic methods for phenomena evolving over time and space. Related to this is the task of deriving algorithms that can be used to learn the unknown model parameters from the measured data. The position involves collaboration within the NEST project partner groups of Johan Elf and Magda Bienko.
The position can include teaching up to 20% depending on availability and interest.
