People

Alkis Kalavasis joined us at Yale as an FDS Postdoctoral Fellow in September 2023, the first year of our postdoc program. Alkis earned his PhD in Computer Science at the National Technical University of Athens (NTUA), Greece in June 2023. Before that, he was an undergraduate student in the School of Electrical and Computer Engineering at the NTUA between 2014 and 2019. His research focuses on theoretical foundations of Machine Learning and their interplay with Statistics & High-Dimensional Probability, Optimization and Computational Complexity. Currently, he is interested in questions regarding various topics such as: (i) Reliable Machine Learning, where the goal is to understand how to perform statistical analysis from biased and corrupted data and how to design algorithms satisfying societal desiderata such as privacy, replicability and fairness; (ii) Statistical Learning Theory, where he aims to understand statistical limits of natural Machine Learning tasks; (iii) Statistical-Computational Trade-offs, where one seeks to detect gaps in problems arising in high-dimensional statistics between what is achievable statistically and what is achievable with known computationally efficient algorithms; (iv) Complexity of Optimization, where he studies the optimization landscape of important Deep Learning problems and mainly aims to discover algorithmic barriers for such tasks.

Edit profile