Project: “BigBayesUQ” – The missing mathematical story of Bayesian uncertainty quantification for big data - Horizon Europe Call: ERC-2022-StG.
Grant Agreement n. 101041064
Research Project:
Recent years have seen a rapid increase in information. This has created an urgent need for fast statistical and machine learning methods that can scale up to big data sets. Standard approaches including the now routinely applied Bayesian methods are becoming computationally infeasible especially in complex models with many parameters and large data sizes. A variety of algorithms have been proposed to speed up these procedures, but these are typically black box methods with very limited theoretical underpinning. In fact, empirical evidence shows the potentially bad performance of such approaches. This is especially concerning in real-world applications e.g. in medicine or astronomy. In this project we aim to provide theoretical guarantees and limitations for such scalable (Bayesian) methods and based on the theoretical understanding derive more accurate procedures.
Job Description:
Research in mathematical statistics focusing on uncertainty quantification for scalable statistical methods. The candidate is expected to publish paper(s), present the work on conference(s) and attend the departmental research seminar in a regular basis. Successful candidates will have to prove strong analytical skills, experience in mathematical statistics (in any of following topics - Bayesian statistics, inverse problems, high-dimensional models, scalable computation, multiple testing, neural networks), and excellent command/highly proficiency in spoken and written English. Good programming skills would be an asset. Experience with scalable Bayesian methods, frequentist analysis of Bayesian approaches and Gaussian processes is an advantage.
If you are interested, please apply at here by 15th January 2025.
