Luc Rey-Bellet

Luc Rey-Bellet is a Professor and Honors Coordinator in the Department of Mathematics and Statistics at the University of Massachusetts Amherst. He has been a faculty member at UMass since 2002, progressing from Assistant Professor to Associate Professor in 2008, and to Full Professor in 2013. His office is located in LGRT 1423K, and he maintains regular office hours on Tuesdays and Fridays.

Dr. Rey-Bellet received his Dipl. Phys. from Eidgenössiche Technische Hochschule Zürich (ETH Zurich) in 1994 and his Ph.D. in Mathematics from Université de Genève in 1998. Following his doctoral studies, he held postdoctoral positions at Rutgers University (1998-1999) and served as a Whyburn Instructor at the University of Virginia (1999-2002) before joining UMass Amherst.

Professor Rey-Bellet's research spans statistical mechanics, applied probability, and their applications across various domains. His work focuses on both theoretical foundations and practical applications, with particular emphasis on non-equilibrium systems, large deviations theory, and computational methods. He has made significant contributions to understanding physical and mathematical properties of non-equilibrium steady states, developing coarse-graining strategies for complex systems, and creating numerical schemes for lattice spin systems and stochastic processes. His research also extends to evolutionary game theory, mathematical economics, Monte-Carlo methods, information theory, uncertainty quantification, and machine learning.

Recent publications reveal a strong trend toward interdisciplinary work at the intersection of probability theory, statistical mechanics, and machine learning. Rey-Bellet has been particularly active in developing mathematical frameworks for generative modeling, with focus on Wasserstein distances, divergence measures, and gradient flows. His work on structure-preserving generative models, group-invariant networks, and uncertainty quantification demonstrates how classical statistical mechanics concepts can inform modern machine learning theory. The consistent theme across his recent work is developing rigorous mathematical foundations for understanding complex probabilistic systems and their computational representations.

Rey-Bellet has secured significant research funding throughout his career, with grants totaling $101K in 2003, $106K in 2006, $99K in 2010, $280K in 2015, $370K in 2020, and $300K in 2023. Notably, he was awarded larger collaborative grants of $900K in 2019, $1.950M in 2021, and $900K in 2016, reflecting the significance and collaborative nature of his research.

While specific teaching awards aren't detailed in the available information, his faculty profile notes "Award-winning teaching," suggesting recognition for his pedagogical contributions. His role as Honors Coordinator further indicates his commitment to undergraduate education and academic excellence.

Professor Rey-Bellet maintains an active research group, frequently collaborating with colleagues including Markos A. Katsoulakis, Jeremiah Birrell, Panagiota Birmpa, and others. His research spans theoretical developments in probability and statistical mechanics while maintaining strong connections to computational methods and applications in machine learning and data science. Current projects appear focused on developing mathematically rigorous frameworks for generative modeling, uncertainty quantification, and understanding the statistical properties of complex systems.