Conferences  >  Mathematics  >  Probability and Statistics, Game Theory  >  United Kingdom

Isaac Newton Institute for Mathematical Sciences, Cambridge

Graphical models have been shown to be a powerful tool for addressing a wide range of problems, including identification of statistical parameters under confounding, selection effects, and missing data, the study of social network data, genomics, and analyses of systems of entangled particle state systems arising in quantum theory. This workshop will bring together methodological and applied researchers working on the theory of graphical models, and their applications. The goal of the workshop is to both exhibit recent advances in the theory of graphical models and to showcase existing connections to applications, as well as to discover new ones.

Isaac Newton Institute for Mathematical Sciences, Cambridge

This workshop explores challenging applications of causal inference methodology in biomedical research. These include a variety of topics including; clinical trials with the question of suitable causal estimands in view of intercurrent events, the application of causal discovery in epidemiology, advances in quantitative bias analyses and methods of Mendelian randomisation with time-varying exposures. The time-dependent nature of real-world biomedical data, with its special sources of potential bias, requires tailored statistical methods and will be a core aspect of the workshop. The aim is to move beyond the common simple settings of binary point treatments towards methods allowing for realistically complex causal questions taking the practical limitations of typical biomedical data into account. The workshop will highlight cutting-edge developments and foster discussion on future directions in this arguably most important field of application of causal inference.

Applied Mathematics (in general),    Clinical Data Management, e-Health

Isaac Newton Institute for Mathematical Sciences, Cambridge

This workshop explores recent advances in the use of flexible machine learning techniques alongside semiparametric and nonparametric statistical methods in causal inference. Recent methodological work has focused on combining modern machine learning tools with the inferential rigor of semiparametric and nonparametric frameworks to estimate causal parameters in complex, high-dimensional settings. The aim is to move beyond the predictive focus typical of standard machine learning, and instead develop estimators that enable valid causal inference while achieving desirable statistical properties such as efficiency and robustness. The workshop will highlight cutting-edge developments and foster discussion on future directions in this rapidly evolving area.

Isaac Newton Institute for Mathematical Sciences, Cambridge

Isaac Newton Institute for Mathematical Sciences, Cambridge

Optimization Theory, Operational Research,    Applied Mathematics (in general)