Sebastian Weichwald, postdoc

In Sebastian's own words, his research aims at closing the gap between statistical causal inference and its fruitful application. The key motivating observation is, that applications are hurdled by the lack of understanding which variables/representations constitute meaningful causal relata and that causal statements on the level of the observed variables are often non-sensical.

Sebastian will work with Jonas Peters and the Copenhagen Causality Lab (CoCaLa) and focus on conceptual/statistical contributions that clarify how causal reasoning within different classes of causal models is being enabled/preserved/impaired upon transforming the observed variables.

His teaching will highlight causality and the intricacies that arise when applying the theoretical concepts to real-world applications.

Sebastian completed his PhD at the Max Planck Institute for Intelligent Systems, Tübingen, with B Schölkopf and at ETH Zurich with PL Bühlmann and JM Buhmann. His thesis on "Pragmatism and Variable Transformations in Causal Modelling" presents examples of putting causality to use in neuroimaging as well as lays out the foundation for his future methodological research.

Sebastian holds degrees from the University of Tübingen (BSc Mathematics) and from UCL (MSc Computational Statistics and Machine Learning) where he completed his master's project on "Causal Effect Recovery from Linear Mixtures" with A Gretton at the Gatsby Computational Neuroscience Unit and graduated with distinction.

Sebastian invites you to drop by his office (04.3.01), send him an email (swe@math.ku.dk) or visit his homepage (https://sweichwald.de) - “if you wish to learn more about my work or optimisation on manifolds in python using Pymanopt, collaborate, or chat about speciality filter coffee”.