An adaptive experimental design for Global sensitivity analysis (GSA) and uncertainty quantification (UQ): Application to groundwater modeling [Slides]
Global sensitivity analysis (GSA) and uncertainty quantification (UQ) in groundwater modeling are challenging because of the significant computational requirements. The computationally efficient surrogate method is an alternative solution for this challenge. This week we will introduce a newly proposed adaptive experimental design strategy for efficient global surrogate construction which aims to improve the computational efficiency in the GSA and UQ for groundwater models.
van der Herten, J., Couckuyt, I., Deschrijver, D., & Dhaene, T. (2015). A fuzzy hybrid sequential design strategy for global surrogate modeling of high-dimensional computer experiments. SIAM Journal on Scientific Computing, 37(2), A1020-A1039.
A critical review of State space models [Slides]
A review of state-space models (SSM) has been carried out. Specific attention has been paid on inference and learning algorithms. Different applications of SSMs have also been discussed.
K. Murphy, Machine Learning - A Probabilistic Perspective, Chapter 18, The MIT Press, 2013.