Data Science Seminar

Active Invariant Causal Prediction: Experiment Selection through Stability

One fundamental difficulty of causal learning is that causal models can generally not be fully identified based on observational data only. Interventional data, that is, data originating from different experimental environments, improves identifiability. However, the improvement depends critically on the target and nature of the interventions carried out in each experiment. Since in real applications experiments tend to be costly, there is a need to perform the right interventions such that as few as possible are required. In this talk, I will present new active learning (i.e. experiment selection) strategies based on causal invariance, which select intervention targets that quickly reveal the direct causes of a response variable of interest in the causal graph.

Biosketch Christina Heinze-Deml

Christina Heinze-Deml is a senior postdoc and lecturer at the Seminar for Statistics at ETH Zurich. During her PhD she was advised by Nicolai Meinshausen and Jonas Peters and she also spent some time at Facebook AI Research and DeepMind.
Among other things, Christina has worked on privacy-preserving distributed machine learning and causality. Within the field of causality, she has been interested in causal structure learning when data sets from different environments are available and recently, she has used a causal framework to make classifiers more robust to certain adversarial domain shifts. More generally, Christina is particularly interested in exploring the connections between causal inference and robust machine learning.

to top
powered by webEdition CMS