Project description
Updating conditional probability approaches
Bayes' theorem developed in the 18th century describes conditional probability, the likelihood of an outcome occurring based on a previous outcome. So-called Bayesian methods in probability and statistics are increasingly used in many areas of basic and applied research. Bayesian approaches assign probability distributions rather than discrete numbers to events or outcomes based on previous observations. Inherently, they include probability distributions and uncertainty associated with both the inputs and the outputs. As the available data with which to develop the models increases exponentially, simpler and more scalable methods are needed. The EU-funded GTBB project will tackle these theoretical and computational challenges to enhance our predictive abilities in fields from neuroscience to security.
Objective
In the modern era of complex and large data sets, there is stringent need for flexible, sound and scalable inferential methods to analyse them. Bayesian approaches have been increasingly used in statistics and machine learning and in all sorts of applications such as biostatistics, astrophysics, social science etc. Major advantages of Bayesian approaches are: their ability to model complex models in a hierarchical way, their coherency and ability to deliver not only point estimators but also measures of uncertainty from the posterior distribution which is a probability distribution on the parameter space at the core of all Bayesian inference. The increasing complexity of the data sets raise huge challenges for Bayesian approaches: theoretical and computational. The aim of this project is to develop a general theory for the analysis of Bayesian methods in complex and high (or infinite) dimensional models which will cover not only fine understanding of the posterior distributions but also an analysis of the output of the algorithms used to implement the approaches.
The main objectives of the project are (briefly):
1. Asymptotic analysis of the posterior distribution of complex high dimensional models
2. Interactions between the asymptotic theory of high dimensional posterior distributions and computational complexity.
We will also enrich these theoretical developments by 3 strongly related domains of applications, namely neuroscience, terrorism and crimes and ecology.
Fields of science
- natural sciencesbiological sciencesneurobiology
- natural sciencesmathematicsapplied mathematicsstatistics and probabilitybayesian statistics
- natural sciencesbiological sciencesecology
- natural sciencesphysical sciencesastronomyastrophysics
- natural sciencescomputer and information sciencesartificial intelligencemachine learning
Programme(s)
Topic(s)
Funding Scheme
ERC-ADG - Advanced GrantHost institution
75775 Paris
France