Learn about
Uncertainty Quantification
This page covers Uncertainty Quantification (UQ) broadly:
from forward to inverse UQ, probabilistic and non-probabilistic approaches, Probabilistic Machine Learning (although we often refer to it with the less fancier names: Bayesian inference), fundamentals of sampling strategies, and also our recent works on UQ.
Codes are shared whenever possible.
from forward to inverse UQ, probabilistic and non-probabilistic approaches, Probabilistic Machine Learning (although we often refer to it with the less fancier names: Bayesian inference), fundamentals of sampling strategies, and also our recent works on UQ.
Codes are shared whenever possible.
Note: the page is under development - last updated: 08/2022
Useful online resources
Many excellent videos on a broad range of UQ topics by Pete Green - link to his YouTube channel and his GitHub page
to be updated - feel free to suggest others!
DVU group resources
A very brief intro to Uncertainty Quantification
Otherwise known as: trying to fit in everything you think is important in 42 slides! Please note that it is a very biased view.
Otherwise known as: trying to fit in everything you think is important in 42 slides! Please note that it is a very biased view.
Cicirello_Intro_to_UQ_082022.pdfSlides on fundamental UQ topics
- (A very brief) intro to Sampling and Monte Carlo Simulations
Intro_to_sampling.pdf- Sampling methods: Inverse Transform Sampling
Sampling_methods_07_2022.pdf
- Sampling methods: Box-Mueller algorithm
Sampling method_box_mueller_07_2022.pdf
- Sampling methods: Accept/reject algorithm
accept_reject_sampling.pdf
Tutorials (with codes)
Learning Sampling Methods for solving Bayesian Model Updating Problems with a guided tutorial with Matlab examples!
Paper: Lye A., Cicirello A., Patelli E., Mechanical System and Signal Processing, 2021, https://doi.org/10.1016/j.ymssp.2021.107760
Full text (preprint) freely available here
Matlab codes available here
Research seminars on UQ topics
11/03/2022 - DDPS online seminar | Interpretable, Explainable and Non-Intrusive Uncertainty Propagation by Alice Cicirello
This talk mainly covers how to propagate efficiently interval uncertainty by exploiting a statistical machine learning tool: Bayesian Optimization.
The paper associated with this talk is:
Cicirello A., Giunta F., Machine Learning based optimization for interval uncertainty propagation. Mechanical System and Signal Processing, 2022
Open access: https://doi.org/10.1016/j.ymssp.2021.108619
Repositories with codes developed to learn about Uncertainty Quantification
DVU recent research publications on Uncertainty Quantification
Igea F., Cicirello A.,Cyclical Variational Bayes Monte Carlo for Efficient Multi-Modal Posterior Distributions Evaluation, Mechanical System and Signal Processing, 2022
https://doi.org/10.1016/j.ymssp.2022.109868Igea F., Chatzis M.N.., Cicirello A., On the Combination of Random Matrix Theory with Measurements On a Single Structure. ASME J. Risk Uncertainty Part B, 2022
https://doi.org/10.1115/1.4054172Cicirello A., Giunta F., Machine Learning based optimization for interval uncertainty propagation. Mechanical System and Signal Processing, 2022
Open access: https://doi.org/10.1016/j.ymssp.2021.108619Lye A., Cicirello A., Patelli E., An efficient and robust sampler for Bayesian inference: Transitional Ensemble Markov Chain Monte Carlo, Mechanical System and Signal Processing, 2022
https://doi.org/10.1016/j.ymssp.2021.108471
Matlab codes available hereLye A., Cicirello A., Patelli E., Sampling Methods for solving Bayesian Model Updating Problems: A Tutorial, Mechanical System and Signal Processing, 2021.
https://doi.org/10.1016/j.ymssp.2021.107760
Matlab codes available here
PAGE UNDER DEVELOPMENT