Uncertainty in multivariate, non-Euclidean, and functional spaces: theory and practice workshop

Introduction

The Department of Statistics, LSE is leading the organization of the Uncertainty in multivariate, non-Euclidean, and functional spaces: theory and practice workshop. The event will be held in the Isaac Newton Institute May 06-09, 2025, and will focus on data science relying on various complex data types (beyond the classical case of vectors) including graphs, rankings, manifolds, time series, sets, probability measures, functions, or persistence diagrams, which have found numerous successful applications with widespread social impact. It will explore (i) the design of computationally tractable, scalable and principled approaches, and the understanding of their statistical-computational trade-offs, (ii) leveraging the intrinsic data structure (e.g., invariant or multi-scale nature, different notions of similarity), (iii) representing and quantifying the uncertainty of predictive models, and (iv) investigating the efficiency of the developed techniques in novel applications.