MS Title:
AI for uncertainty quantification
 

Description:
Uncertainty quantification (UQ) involves quantitatively characterizing all sources of uncertainties arising from both computational and real-world applications. It plays a pivotal role in various scientific and engineering domains, particularly in situations where decisions or product designs hinge on imperfectly known system aspects due to a lack of information or intrinsic randomness. A comprehensive UQ framework includes many sub-tasks such as uncertainty characterization, forward uncertainty propagation, inverse uncertainty propagation, uncertainty sensitivity analysis, etc. All the subtasks of UQ pose great challenges in numerical computation.

Artificial intelligence (AI) including machine learning is the scientific study of algorithms and statistical models that allow computers to learn from existing data without being explicitly programmed. In recent years, the application of AI in a wide range of industries has grown rapidly. Hence, it has brought new hopes for addressing UQ challenges. However, the recent developments in this area are far from mature for solving all the above-mentioned tasks. The aim of this mini-symposium is to collect the latest developments in the realm of AI for UQ, offering a platform to explore innovative approaches and solutions across all facets of uncertainty quantification. 

Specific contributions related to both methodology developments and engineering applications regarding but not restricted to following aspects are welcome:

• Big data-based engineering loading modeling.
• New surrogate modeling techniques tailored to some computationally-demanding problems.
• Efficient reliability analysis methods to challenging problems.
• Efficient sensitivity analysis techniques.
• Time history predictions and time-variant reliability analysis of complex dynamic problems.
• Reliability-based design optimization using advanced algorithms.
• Physics-informed neural networks-based solutions.
• Model update with structural health monitoring.

Session Chairs:
Tong Zhou, The Hong Kong Polytechnic University, Hong Kong, China, Email: tong-cee.zhou@polyu.edu.hk
Chao Dang, Leibniz Universität Hannover, Hannover, Germany, Email: chao.dang@irz.uni-hannover.de
Yongbo Peng, Tongji University, Shanghai, China, Email: pengyongbo@tongji.edu.cn
Michael Beer, Leibniz Universität Hannover, Hannover, Germany, Email: beer@irz.uni-hannover.de
Bruno Sudret, ETH Zurich, Zurich, Switzerland, Email: sudret@ethz.ch
Enrico Zio, Politecnico di Milano, Milan, Italy. Email: enrico.zio@polimi.it