Kilbertus team
RELIABLE MACHINE LEARNING
BUILDING ROBUST, THEORETICALLY SUBSTANTIATED AND SOCIALLY BENEFICIAL MACHINE LEARNING SYSTEMS
Our Helmholtz AI young investigator group broadly investigates machine learning systems that interact with humans, for example by making consequential decisions, affecting our behavior, or challenging our privacy. We focus on reliable, fair, and privacy preserving algorithms for these settings. As a key tool for trustworthy machine learning, we also aim at making causal inference techniques more applicable to systems involving humans.
Research lines
- Causality
- Reliable ML
- Methodology