SIAM Conference on Uncertainty Quantification (UQ26)

About the Conference

This is the conference of the SIAM Activity Group on Uncertainty Quantification. 

This conference is being held in cooperation with the American Statistical Association (ASA).

Uncertainty quantification (UQ) is essential for establishing the reliability of predictions made by computational models. Such models are often deterministic, and the UQ discipline has traditionally focused on how to quantify uncertainty in that case. Statistical methods play a major role in that effort, and represent an alternative modeling strategy when less is known about the system of interest. The two approaches are naturally synergistic, and with the emergence of machine learning as a practical tool, it is becoming more important than ever to view the whole UQ ecosystem through a unifying lens.

UQ is application-driven and inherently interdisciplinary, relying on a broad range of mathematical and statistical foundations, domain knowledge, and algorithmic and computational tools. UQ26 will bring together mathematicians, statisticians, scientists, engineers, and others interested in the theory, development, and application of UQ methods. Major conference themes will include mathematical and statistical foundations, data-driven approaches and computational advances, applications of UQ in biology, medicine, environmental and climate sciences, decision making for societal benefit, and all areas of physical science and engineering. The goal of the conference is to provide a forum for exchanging ideas between diverse groups from academia, industry, and government laboratories, thereby enhancing communication and contributing to future advances in the field.

Connect with other attendees on LinkedIn. 

Included Themes

Major Themes

• Computational advances
• Data-driven methods
• Machine learning for UQ and UQ for machine learning
• Mathematical and statistical foundations of UQ
• UQ for decision making and societal benefit
• UQ in biosciences, bioengineering, and biomedicine
• UQ in climate and environmental sciences
• UQ in engineering
• UQ in physical sciences

Subtopics

• Adversarial and generative approaches to UQ
• Causal inference and graphical models
• Combining physical and statistical modeling
• Data assimilation, data fusion and control
• Design of experiments
• High-dimensional approximation and dimension reduction
• Infinite-dimensional analysis and approximation
• Inverse problems
• Model discrepancy
• Monte Carlo and quasi-Monte Carlo methods
• Multiscale, multilevel and multifidelity UQ methods
• Optimization for inverse problems and variational inference
• Optimization for machine learning
• Physics-informed machine learning
• Rare and extreme events
• Reduced-order models
• Remote sensing
• Risk analysis based on UQ
• Risk-averse and robust optimization
• Spatial statistical methods
• Surrogate models, emulators and Gaussian process methods
• UQ for complex and coupled systems
• UQ for digital twins
• UQ for spatial and spatio-temporal processes
• UQ-informed policy and decisions
• Verification and validation