Stephan Mandt

Stephan Mandt is an Associate Professor of Computer Science and Statistics at the University of California, Irvine. He leads research at the intersection of deep generative modeling, uncertainty quantification, neural data compression, and AI-driven scientific discovery. Previously, he directed the machine learning group at Disney Research and held postdoctoral positions at Princeton and Columbia University. His work is supported by NSF, DARPA, IARPA, DOE, Disney, Intel, and Qualcomm. He has received the NSF CAREER Award, UCI Mid-Career Excellence Award, and German Research Foundation's Mercator Fellowship.

Education: PhD in Theoretical Physics from the University of Cologne (German National Merit Scholarship recipient). Served as Program Chair for AISTATS 2024, General Chair for AISTATS 2025, and Action Editor for JMLR and Transactions on Machine Learning Research. Active in organizing tutorials at NeurIPS, AAAI, and UAI.

Research Interests: Focuses on generative AI foundations, including diffusion models, variational autoencoders, and neural compression techniques. Explores applications in climate science, physics, and chemistry. Current projects include AI for scientific discovery, resource-efficient learning algorithms, and uncertainty-aware neural networks.

Grants & Awards: Recognized as a Chan Zuckerberg Investigator and AI Resident. Holds the Kavli Fellowship from the U.S. National Academy of Sciences. His research bridges machine learning theory with interdisciplinary scientific challenges, emphasizing scalable and interpretable algorithms.

Labs & Teams: Leads the Mandt Lab at UCI, supervising PhD students and postdocs in generative AI and scientific machine learning. Collaborates with institutions like ETH Zurich (as Guest Professor) and the Chan Zuckerberg Initiative.