Recomendamos que instale un navegador más moderno como por ejemplo Mozilla Firefox, Microsoft Edge (Windows 10), Vivaldi, Brave, Opera o Google Chrome. Este sitio web también es compatible con Safari en macOS.
I obtained my PhD in Applied Mathematics and Statistics from The University of Nottingham in 2015. During my thesis, I developed various statistical and computational techniques for performing uncertainty quantification in computationally expensive models of carbon capture and storage processes.
I have an established interest in the relationship between machine learning and mathematical models, in particular for those involving environmental fluid mechanics. My primary area of research has been the development of new techniques for uncertainty quantification in stochastic models, these include the use of linear (e.g., principal component analysis) and non-linear (e.g., manifold learning) dimension reduction methods defined in high-dimensional input and output spaces, and the development of statistical surrogates for computationally expensive numerical simulators, with particular focus on Gaussian process emulators. I have also investigated multilevel Monte Carlo and quasi-Monte Carlo methods for uncertainty quantification in flow and transport in random heterogeneous porous media.
In the past, I have worked as a postdoctoral researcher at the University of Warwick. Withing the Statistical Department, I was involved in the investigation of spatial arrangements of dysfunctional pixels in Computed Tomography by studying point patterns, complete spatial randomness, clustering and regularity.
I also worked during two years in a research centre called LUCIDEON, based on Stoke-on-Trent (UK), where as part of the Simulation and Modelling Department I was working on the numerical simulation of powder compaction and dead-end filtration processes by using computational modelling covering both Finite Element and fluid dynamic techniques.
My recent investigations have been focused on investigating the application of machine learning to the prediction of crowd and traffic motion in modern and future urban environments. More precisely, I worked on developing and applying appropriate machine learning and uncertainty quantification techniques to a) make city-scale simulation computationally realisable; b) use real-time data to improve predictive capability of crowd/traffic models and to quantify uncertainty in model outputs.