Alireza Doostan, University of Colorado Boulder
Alexandre Cortiella, University of Colorado Boulder
Assad Oberai, University of Southern California
Jianxun Wang, University of Notre Dame
Recent advances in data acquisition systems along with modern data science and machine learning techniques have fostered the development of accurate data-driven approaches, such as inverse modeling for model calibration and system identification, in science and engineering fields. In particular, system identification, i.e., deducing accurate mathematical models from measured observations, is key to improved understanding of complex phenomena, dominant feature analysis, design of experiments, and system monitoring and control. Furthermore, the emergence of multi-fidelity approaches provides further roles for data-driven models as a low-cost model evaluation for UQ tasks.
This mini-symposium focuses on recent developments in discovering governing equations, deriving discrepancy terms, and building reduced order models of non-linear dynamics from simulation or experimental data. Some of the techniques include sparse regression, physics-based deep learning, operator inference, and reinforcement learning. Of particular interest are data-driven methods addressing challenges regarding measurement noise, sampling strategies, identifiability, and scalability for accurate and robust model extraction in complex non-linear dynamical systems. Additionally, application of the extracted models to visualization, data compression/reconstruction, real-time controls, or uncertainty quantification are encouraged.