The principles and methodology of causal modeling will significantly enhance the robustness, foundation, and interpretability of the AI cycle (from data to models, from models to decisions, and from decisions to new data). Within this framework, we will focus on several key aspects. Initially, we will prioritize the development of a causal representation. Current approaches often involve learning unsupervised representations that reduce dimensionality but sacrifice the interpretability of variables. We aim to address this issue by considering interpretable meta-variables. Furthermore, we aim to investigate the relationship between feature selection in statistics and causal inference. For instance, we plan to compare the features selected by the Lasso estimator and the Markov blanket associated with a causal graph. Can we combine both? Which is preferable for robust prediction? Finally, we intend to conduct an in-depth analysis of robustness in relation to out-of-distribution scenarios. This will involve a theoretical assessment of the generalization error on a test set that is significantly different from the training set.
ACTIVITIES
1 postdoc (Daria Bystrova) and 1 PhD student (Théotime Le Goff) starting in November 2023
A visit to the Copenhagen Causality Lab in the spring of 2024
Published on October 13, 2026 Updated on October 13, 2026
Share the linkCopyCopiedClose the modal windowShare the URL of this pageI recommend this page:Available at this address:The page will then be accessible from your "My Favorites" menu.Stop the videoPlay the videoMutePlay audioChat: Have a question?Robo Fabrica ChatbotMatomo Traffic StatisticsX (formerly Twitter)
