ML-Talk am 18.06.2026: Prof. Stephan Mandt (University of California, Irvine), "Generative AI for Geospatial Data"


  • Event Location: Villa Denis, Diemerstein 9, 67468 Frankenstein

    Event Date: 18.06.2026

    Contact Person: Prof. Dr. Marius Kloft

    https://ml.cs.rptu.de/

    Logo ML-Talk am 18.06.2026: Prof. Stephan Mandt (University of California, Irvine), "Generative AI for Geospatial Data"

    The Machine Learning Group of Prof. Marius Kloft
    at the Department of Computer Science Department at RPTU Kaiserslautern-Landau
    cordially invites you to a talk on

        "Generative AI for Geospatial Data"
        by Prof. Stephan Mandt (University of California, Irvine)

    on

        Thursday, June 18th 2026 at 12:30 in Villa Denis

    Abstract:

    Many problems in the natural sciences involve complex, high-dimensional spatiotemporal data, from scientific imaging to weather and climate systems.
    While diffusion models have emerged as powerful tools for modeling such data, their standard formulations are primarily designed for natural images and videos and do not directly address the physical constraints and observational consistency required in scientific applications.
    This talk discusses diffusion models as tools for scientific inference, with a focus on inverse problems. I will present guidance methods for incorporating observational constraints in both static and dynamical settings, and show applications to weather and climate data,
    including heavy-tailed precipitation modeling and spatiotemporal diffusion models for probabilistic downscaling and forecasting. These examples illustrate how generative AI can move beyond realistic sample generation toward uncertainty-aware inference in complex physical systems.

    Bio:

    Stephan Mandt is a Professor of Computer Science and Statistics at the University of California, Irvine, Director of the UCI Center for Machine Learning, and co-director of UCI’s AI in Science Institute.
    His research contributes to the foundations and applications of generative AI, with a focus on diffusion models, uncertainty quantification, neural compression, resource-efficient inference, and AI-driven scientific discovery.
    He is a Chan Zuckerberg Investigator and AI Resident, NSF CAREER awardee, and Kavli Fellow.

    This is a scientific talk which will be held in English.
    The AI Alliance Rhineland-Palatinate supported the organization of the event.
    -- Steffen Reithermann

    Prof. Stephan Mandt - University of California, Irvine (UCI)

    Prof. Stephan Mandt - University of California, Irvine (UCI)