Seminar series

This page brings together selected seminars from the Erlangen Hub, featuring invited speakers and research talks.

Browse seminars

Branton DeMoss, University of Oxford
Geometry, Complexity, and Generalization in Learning Systems
In this talk, Branton DeMoss (University of Oxford) will discusses the relationship between compression, complexity, and the geometry of the loss landscape.

Tristan Madeleine, University of Southampton
Learning on graphs: expressivity challenges from graph pooling
Graphs provide a natural framework for relational data across many domains, presenting interesting practical and theoretical challenges for machine learning.

Eng-Jon Ong, QMUL
Estimating Intrinsic Dimensionality with L2N2 This talk introduces L2N2, a simple yet powerful ID estimator based on nearest-neighbour distance ratios that achieves state-of-the-art performance with minimal computational overhead.

Iolo Jones, University of Oxford
Computing Diffusion Geometry
Diffusion geometry is a new theory that reformulates classical calculus and geometry in terms of a diffusion process, allowing these theories to generalise beyond manifolds and be computed from data.

Alessandro Micheli, Imperial
Riemannian Neural Optimal Transport
This talk introduces Riemannian Neural Optimal Transport (RNOT), a continuous neural parameterisation of OT maps that avoids discretisation and incorporates geometric structure directly.

Ambrose Yim, University of Oxford
Geometry of Loops on the Möbius Band
This talk proposes propose a novel representation of a loop’s geometry by representing the distance matrix of the loop as a Morse function on a Möbius band.