SubsidieMeestersThe missing mathematical story of Bayesian uncertainty quantification for big data
This project aims to enhance scalable Bayesian methods through theoretical insights, improving their accuracy and acceptance in real-world applications like medicine and cosmology.
Projectdetails
Introduction
Recent years have seen a rapid increase in available information. This has created an urgent need for fast statistical and machine learning methods that can scale up to big data sets.
Challenges in Current Methods
Standard approaches, including the now routinely used 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 support.
Concerns in Real-World Applications
In fact, empirical evidence shows the potentially bad performance of such methods. This is especially concerning in real-world applications, e.g., in medicine.
Project Goals
In this project, I shall open up the black box and provide a theory for scalable Bayesian methods combining recent, state-of-the-art techniques from Bayesian nonparametrics, empirical process theory, and machine learning.
Focus Areas
I focus on two very important classes of scalable techniques:
- Variational Bayes
- Distributed Bayes
Establishing Guarantees and Limitations
I shall establish guarantees, but also limitations, of these procedures for estimating the parameter of interest, and for quantifying the corresponding uncertainty, within a framework that will also convince outside of the Bayesian paradigm.
Expected Outcomes
As a result, scalable Bayesian techniques will have more accurate performance, and also better acceptance by a wider community of scientists and practitioners.
Nature of the Research
The proposed research, although motivated by real-world problems, is of a mathematical nature. In the analysis, I consider mathematical models, which are routinely used in various fields (e.g., high-dimensional linear and logistic regressions are the workhorses in econometrics or genetics).
Practical Applications
My theoretical results will provide principled new insights that can be used, for instance, in multiple specific applications I am involved in, including:
- Developing novel statistical methods for understanding fundamental questions in cosmology
- Early detection of dementia using multiple data sources.
Financiële details & Tijdlijn
Financiële details
| Subsidiebedrag | € 1.492.750 |
| Totale projectbegroting | € 1.492.750 |
Tijdlijn
| Startdatum | 1-8-2022 |
| Einddatum | 31-7-2027 |
| Subsidiejaar | 2022 |
Partners & Locaties
Projectpartners
- UNIVERSITA COMMERCIALE LUIGI BOCCONIpenvoerder
Land(en)
Vergelijkbare projecten binnen European Research Council
| Project | Regeling | Bedrag | Jaar | Actie |
|---|---|---|---|---|
Provable Scalability for high-dimensional Bayesian LearningThis project develops a mathematical theory for scalable Bayesian learning methods, integrating computational and statistical insights to enhance algorithm efficiency and applicability in high-dimensional models. | ERC Starting... | € 1.488.673 | 2023 | Details |
Advanced Numerics for Uncertainty and Bayesian Inference in ScienceANUBIS aims to enhance quantitative scientific analysis by unifying probabilistic numerical methods with machine learning and simulation, improving efficiency and uncertainty management in data-driven insights. | ERC Consolid... | € 1.997.250 | 2024 | Details |
Scalable Learning for Reproducibility in High-Dimensional Biomedical Signal Processing: A Robust Data Science FrameworkScReeningData aims to develop a scalable learning framework to enhance statistical robustness and reproducibility in high-dimensional data analysis, reducing false positives across scientific domains. | ERC Starting... | € 1.500.000 | 2022 | Details |
Inference in High Dimensions: Light-speed Algorithms and Information LimitsThe INF^2 project develops information-theoretically grounded methods for efficient high-dimensional inference in machine learning, aiming to reduce costs and enhance interpretability in applications like genome-wide studies. | ERC Starting... | € 1.662.400 | 2024 | Details |
High-dimensional nonparametric Bayesian causal inferenceDevelop Bayesian nonparametric methods for high-dimensional causal inference to enhance variable selection and uncertainty quantification, enabling reliable causal conclusions across various fields. | ERC Starting... | € 1.499.770 | 2023 | Details |
Provable Scalability for high-dimensional Bayesian Learning
This project develops a mathematical theory for scalable Bayesian learning methods, integrating computational and statistical insights to enhance algorithm efficiency and applicability in high-dimensional models.
Advanced Numerics for Uncertainty and Bayesian Inference in Science
ANUBIS aims to enhance quantitative scientific analysis by unifying probabilistic numerical methods with machine learning and simulation, improving efficiency and uncertainty management in data-driven insights.
Scalable Learning for Reproducibility in High-Dimensional Biomedical Signal Processing: A Robust Data Science Framework
ScReeningData aims to develop a scalable learning framework to enhance statistical robustness and reproducibility in high-dimensional data analysis, reducing false positives across scientific domains.
Inference in High Dimensions: Light-speed Algorithms and Information Limits
The INF^2 project develops information-theoretically grounded methods for efficient high-dimensional inference in machine learning, aiming to reduce costs and enhance interpretability in applications like genome-wide studies.
High-dimensional nonparametric Bayesian causal inference
Develop Bayesian nonparametric methods for high-dimensional causal inference to enhance variable selection and uncertainty quantification, enabling reliable causal conclusions across various fields.