Second USACM Thematic Conference on Uncertainty Quantification for Machine Learning Integrated Physics Modeling (UQ-MLIP 2024)

Computational models of real world systems are increasingly integrating data-driven models from the field of machine learning with physics models based on first-principles. It is thus of greatest importance to carefully characterize and quantify the uncertainties associated with each model class. Furthermore, the propagation of uncertainties to the outcomes of the integrated models demands for novel approaches or extension of existing methodologies. Other topics of interest include their applications to related domains such as digital twinning, model reduction, large scale integrated computations, active decision making under UQ frameworks, etc.

The objectives of the conference are to bring together leading experts, scientists, young researchers to exchange about the latest developments on these topics and identify challenges and opportunities to push the field forward.

This conference is being organized by the leadership committees of both the USACM TTA on Data-Driven Modeling and the USACM TTA on Uncertainty Quantification and Probabilistic Modeling.

Organizing Committee

Alireza Doostan, University of Colorado Boulder
Johann Guilleminot, Duke University
Assad Oberai, University of Southern California
Abani Patra, Tufts University
Paris Perdikaris, University of Pennsylvania
Serge Prudhomme, Polytechnique Montréal
Michael Shields, Johns Hopkins University
Jian-Xun Wang, University of Notre Dame