Marco Mondelli

Marco Mondelli is a Professor at the Institute of Science and Technology Austria (ISTA), where he has worked since 2019. His research spans machine learning, high-dimensional statistics, data science, and information theory. Previously, he was an Assistant Professor at ISTA (2019-2025), a Postdoctoral Research Scholar at Stanford University (2017-2019), and a Research Fellow at the Simons Institute for the Theory of Computing (2018). He earned his Ph.D. in Computer and Communication Sciences at EPFL in 2016 under Rüdiger Urbanke.

Research Interests: His work focuses on theoretical aspects of machine learning and high-dimensional statistics, including neural collapse, privacy in over-parameterized models, approximate message passing algorithms, and information limits. He also explores applications in quantitative genetics and wireless communications.

Publications highlight trends in deep learning theory, statistical inference, and generative models. His awards include the ERC Starting Grant (2024), Simons-Berkeley Fellowship (2018), and Best Paper at STOC (2016). He currently advises Ph.D. students in his group and serves on editorial boards for journals like IEEE Transactions on Information Theory.