The 9th David Finney Lecture (2025) We are pleased to announce that the 9th David Finney Lecture will be presented by Prof Laura Sangalli (Politecnico di Milano). You can read more about David Finney as well as watch some of the previous lectures here. Talk details Title: Physics-Informed Statistical Learning for spatial and functional dataAbstract: This lecture presents a family of physics-informed statistical learning methods for spatial and functional data. The models advance classical statistical problems, such as nonparametric and semiparametric regression, by incorporating roughness penalties based on differential operators, ranging from simple Laplacians to problem-specific partial differential equations. These penalties encode knowledge of both the physics underlying the phenomena and the geometry of the observation domain. This modeling feature enables the analysis of data observed on non-convex multidimensional regions (e.g., data scattered over land regions or water bodies with complex shorelines), on curved surfaces (e.g., data observed on sections of the globe; biological signals mapped on the surface of the brain, the heart, or other internal organs), and on linear networks (e.g., data collected over road or river systems).The approach combines the richness and flexibility of regression and maximum likelihood estimation with the powerful modeling capabilities of partial differential equations. This strong synergy, however, gives rise to theoretical and methodological challenges not encountered in classical settings, necessitating tailored arguments to establish large-sample properties and to devise accurate and powerful inferential procedures. Because the considered estimation problems do not admit closed-form solutions, numerical discretization is required; in particular, the use of basis functions defined on unstructured meshes offers exceptional modeling flexibility, enabling the capture of highly localized signals, strong anisotropies, and non-stationary patterns. The practical value of these methods will be illustrated through applications to complex problems in the life and environmental sciences. Schedule: To be announced. Prof Laura Sangalli Laura Sangalli is Professor of Statistics at Politecnico di Milano, Italy, and a member of MOX, the Laboratory for Modeling and Scientific Computing of the Department of Mathematics. Her research focus on statistical methods for complex and high-dimensional data. She serves as panel member for ERC Starting Grants (PE1, Mathematics) and other international funding boards, and is currently Vice-President of GRASPA, the environmental statistics section of the Italian Statistical Society. She currently acts as co-Editor of Statistical Methods and Applications and co-Editor-in-Chief of the Journal of Computational and Graphical Statistics. Prof Sangalli's personal webpage Oct 22 2025 13.30 - 15.30 The 9th David Finney Lecture (2025) The 9th David Finney Lecture will be presented by Prof Laura Sangalli in October 2025. TBC
The 9th David Finney Lecture (2025) We are pleased to announce that the 9th David Finney Lecture will be presented by Prof Laura Sangalli (Politecnico di Milano). You can read more about David Finney as well as watch some of the previous lectures here. Talk details Title: Physics-Informed Statistical Learning for spatial and functional dataAbstract: This lecture presents a family of physics-informed statistical learning methods for spatial and functional data. The models advance classical statistical problems, such as nonparametric and semiparametric regression, by incorporating roughness penalties based on differential operators, ranging from simple Laplacians to problem-specific partial differential equations. These penalties encode knowledge of both the physics underlying the phenomena and the geometry of the observation domain. This modeling feature enables the analysis of data observed on non-convex multidimensional regions (e.g., data scattered over land regions or water bodies with complex shorelines), on curved surfaces (e.g., data observed on sections of the globe; biological signals mapped on the surface of the brain, the heart, or other internal organs), and on linear networks (e.g., data collected over road or river systems).The approach combines the richness and flexibility of regression and maximum likelihood estimation with the powerful modeling capabilities of partial differential equations. This strong synergy, however, gives rise to theoretical and methodological challenges not encountered in classical settings, necessitating tailored arguments to establish large-sample properties and to devise accurate and powerful inferential procedures. Because the considered estimation problems do not admit closed-form solutions, numerical discretization is required; in particular, the use of basis functions defined on unstructured meshes offers exceptional modeling flexibility, enabling the capture of highly localized signals, strong anisotropies, and non-stationary patterns. The practical value of these methods will be illustrated through applications to complex problems in the life and environmental sciences. Schedule: To be announced. Prof Laura Sangalli Laura Sangalli is Professor of Statistics at Politecnico di Milano, Italy, and a member of MOX, the Laboratory for Modeling and Scientific Computing of the Department of Mathematics. Her research focus on statistical methods for complex and high-dimensional data. She serves as panel member for ERC Starting Grants (PE1, Mathematics) and other international funding boards, and is currently Vice-President of GRASPA, the environmental statistics section of the Italian Statistical Society. She currently acts as co-Editor of Statistical Methods and Applications and co-Editor-in-Chief of the Journal of Computational and Graphical Statistics. Prof Sangalli's personal webpage Oct 22 2025 13.30 - 15.30 The 9th David Finney Lecture (2025) The 9th David Finney Lecture will be presented by Prof Laura Sangalli in October 2025. TBC
Oct 22 2025 13.30 - 15.30 The 9th David Finney Lecture (2025) The 9th David Finney Lecture will be presented by Prof Laura Sangalli in October 2025.
