The Shields Uncertainty Research Group (SURG) is the research group of Prof. Michael Shields at Johns Hopkins University. SURG conducts fundamental research in uncertainty quantification (UQ), scientific machine learning (SciML), and probabilistic modeling for engineering systems. Members of SURG have wide-ranging and diverse interest, studying the influence of uncertainties across mathematical and engineering disciplines. These include, but are not limited to:

• Materials modeling from atomistic to structural scales
• Natural hazards modeling including stochastic modeling of seismic and wind hazards.
• Risk and reliability analysis for large-scale structures
• Blast, shock, and impact analysis
• Biomechanics and biomaterials
• Nuclear structures and materials

The group develops new methodologies to model uncertainties, which include:

• Efficient sampling methods for Monte Carlo methods
• Surrogate modeling methods including polynomial chaos expansions and Gaussian process regression
• Scientific Machine Learning models such as neural networks and neural operators
• Reliability and probability of failure analysis
• Methods to quantify imprecise probabilities
• Stochastic simulation methods and generative models for high-dimensional random processes and fields
• Bayesian methods for probabilistic inference
• Manifold learning and dimension reduction.

To learn more, see the following pages