I did my PhD at the intersection of Theoretical Computer Science, AI and Computational Cognitive Science at the University of Bristol. I am now a research scientist at the Max Planck Institute ESI.
My research takes an interdisciplinary approach to interpretability in AI and the Cognitive & Brain Sciences.
I integrate conceptual, empirical, and formal methods (e.g., parameterized complexity theory) to develop and evaluate mathematical and computational models that (1) explain natural and artificial cognition, and (2) characterize the scope and limits (e.g., formal guarantees) of the scientific procedures used to discover such explanations.
A complexity-theoretic view of inner interpretability and its integration with experimental efforts
Demystifying AI Interpretability through Computational Meta-Theory
The Scope and Limits of AI Interpretability for Science and Society