报告人简介：工学博士，现任湖南大学土木工程学院教授、博士生导师，国家级高层次青年人才，湖南大学岳麓学者。先后于2016年和2019年在湖南大学取得学士、硕士学位。于2023年在德国汉诺威莱布尼兹大学取得博士学位，师从国际结构安全与可靠性协会主席Michael Beer教授。博士毕业后，于2024年--2026年在德国多特蒙德工业大学开展博士后研究，合作导师为国际著名SCI期刊RESS和MSSP副主编Matthias Faes教授。长期从事工程结构不确定性分析，在Structural Safety（JCR Q1，IF=5.7）、Reliability Engineering & System Safety （JCR Q1，IF=13.7）、 Mechanical Systems and Signal Processing （JCR Q1，IF=10.2）、Computer Methods in Applied Mechanics and Engineering（JCR Q1，IF=7.6）等领域内国际著名SCI期刊共发表论文约50篇，谷歌学术总引用1600余次, h指数为28。兼任多个英文期刊的青年编委，以及十余本SCI期刊的审稿人。

研究方向：结构可靠性; 结构随机动力学; 模型确认与验证; 贝叶斯数值计算

报告简介：Uncertainty quantification (UQ), in its broadest sense, is the science of identifying, characterizing, and managing uncertainties in both computational models and real-world systems. Forward UQ is a key branch that examines how input uncertainties influence model outputs, with significant applications in applied sciences and engineering. This talk will address forward UQ from a Bayesian active learning perspective, with a focus on time-independent, time-dependent and dynamic reliability analysis, as well as response probability distribution estimation. The core idea is to treat the quantities of interest as Bayesian inference problems and then use the resulting posterior statistics to design the key components of active learning—such as stopping criteria and learning functions. Numerical examples show that these methods can significantly improve computational efficiency while maintaining desired accuracy.