Important classes of functional data, such as random measures and random operators, are intrinsically constrained and need to be represented as elements of suitable non-Euclidean spaces, such as Hilbert manifolds. The interplay between the infinite dimensionality of their variation and the non-linearity of their representation space poses novel challenges for statistics, at the interface of several domains, such as shape analysis, stochastic geometry, manifold learning, and functional data analysis. This workshop will bring together experts from these fields to explore these novel challenges and future perspectives.
Topics: Part of the Semester : Functional Data Analysis