Paper 'What Makes Treatment Effects Identifiable? Characterizations and Estimators Beyond Unconfoundedness' won Best Paper Award at COLT 2025; paper 'On diffusion models and distribution learning' received Short Best Paper Award at ICLR 2025 DeLTa workshop; organized the 'Reliable ML with Unreliable Data' workshop at NeurIPS 2025.
Research Experience
FDS Postdoctoral Fellow at Yale University; previously a PhD student at NTUA, working on related research projects.
Education
FDS Postdoctoral Fellow at Yale University; PhD in Computer Science from National Technical University of Athens (NTUA), supervised by Dimitris Fotakis and Christos Tzamos; undergraduate degree from the School of Electrical and Computer Engineering at NTUA.
Background
Research interests include statistical and computational learning theory. Specifically, he focuses on learning from imperfect data, generative modeling, and the generalization and stability of algorithms.
Miscellany
On the 2025/26 job market; contact email: alkis.kalavasis[at]yale.edu