Tim G. J. Rudner's Research Group

My group's research spans machine learning, AI safety, and AI governance. The goal of our research is to create well-specified, robust, and transparent machine learning models that can be deployed in safety-critical and high-stakes settings. We are particularly interested in (i) understanding and expanding the statistical foundations of generative models (with a focus on robustness to domain shifts, reliable uncertainty quantification, and interpretability), (ii) creating trustworthy AI agents, and (iii) designing regulatory approaches that enable the effective governance of frontier AI models.