Kai Cheng, Ph.D.

Room: N3632 

Phone: +49 89 289 23053


Office hours: by arrangement

Curriculum Vitae

  • Since Nov 2023 | Humboldt  research fellow at the Engineering Risk Analysis Group, Technical University of Munich
  • Sep 2021 - Sep 2023 | Postdoctoral researcher at the Department of Mathematics and Computer Science, University of Southern Denmark. 
  • Apr 2018 - Aug 2021 | PhD student in Fight Vehicle Design, Northwestern Polytechnical University.
  • Sep 2015 - Mar 2018 | M.Sc in Fight Vehicle Design, Northwestern Polytechnical University.
  • Sep 2011 - Jul 2015 | B.Sc in Mechanics, China University of Petroleum (East China)


  • Surrogate model and machine learning
  • Model reduction of dynamical system
  • Bayesian optimization of high-dimensional models
  • Rare event probability estimation and reliability analysis
  • Uncertainty quantification of engineering models

Selected Journal Publications

  • Kai Cheng and Zhenzhou Lu. Adaptive sparse polynomial chaos expansion for global sensitivity analysis based on support vector regression. Computers & Structures, 194:86–96, 2018.
  • Kai Cheng and Zhenzhou Lu. Structural reliability analysis based on ensemble learning of surrogate models. Structural Safety, 83:101905, 2020.
  • Kai Cheng and Zhenzhou Lu. Adaptive bayesian support vector regression model for structural reliability analysis. Reliability Engineering & System Safety, 206:107286, 2021.
  • Kai Cheng, Zhenzhou Lu, Sinan Xiao, and Jingyu Lei. Estimation of small failure probability using generalized subset simulation. Mechanical Systems and Signal Processing, 163:108114, 2022.
  • Kai Cheng, Iason Papaioannou, Zhenzhou Lu, Xiaobo Zhang, and Yanping Wang. Rare event estimation with sequential directional importance sampling. Structural Safety, 100:102291, 2023.
  • Kai Cheng and Ralf Zimmermann. Sliced gradient-enhanced Kriging for high-dimensional function approximation, Accepted for publication in SIAM Journal on Scientific Computing.

Invited Talks

  • Improved gradient enhanced Kriging model for high dimensional function approximation, In15th World Congress on Computational Mechanics & 8th Asian Pacific Congress on Computational Mechanics
  • Gradient-enhanced polynomial chaos expansion for high-dimensional function approximation. In: Proceedings of ICOSSAR 2021-2022, 13th International Conference on Structural Safety & Reliability, Tongji University, Shanghai, China.
  • Rare event estimation with sequential directional importance sampling. 8th European Congress on Computational Methods in Applied Sciences and Engineering (ECCOMAS Congress 2022), Oslo, Norway, June 8, 2022.


  • Stochastic Finite Element Methods  (TA, WS 23/24, 6 ECTS, @TUM