Speakers and mentors

Keynote speakers

Prof Michael Bronstein

Prof Michael Bronstein

Imperial College London, Twitter

Prof Ulrike Tillmann FRS

Prof Ulrike Tillmann FRS

University of Oxford

Speakers

Dr Michael Betancourt

Symplectomorphic LLC

Google Brain

Dr Maximilian Nickel

Facebook AI Research

Prof Martin Rumpf

Universität Bonn

Dr Ruriko Yoshida

Naval Postgraduate School, California

Prof Sanja Fidler

University of Toronto, Vector Institute, Nvidia

Prof Niloy J. Mitra

University College London

Prof Maks Ovsjanikov

École Polytechnique

Prof Gabriele Steidl

Technische Universität Berlin

Prof Marinka Zitnik

Harvard University

Project mentors

Daniele Grattarola

University of Lugano (CH)

Coarsening disassortative graphs

Dr Kelly Spendlove

University of Oxford

Stability or Collapse: Topological Properties of Deep Autoencoders

Qualcomm AI Research

Dr Octavian-Eugen Ganea

MIT CSAIL

Implicit Node and Edge Features for More Expressive Graph Neural Networks

Tel Aviv University

Surface reconstruction from point clouds

MIT CSAIL

Pretraining graph neural networks with ELECTRA

DeepMind

Uncovering and correcting biases in neuroimaging studies

Dr Bamdev Mishra

Microsoft

Manifold optimization and recent applications

Sebastian Neumayer

Technische Universität Berlin

Morphing of manifold-valued images

Dr Shubhendu Trivedi

MIT CSAIL

Efficient Fully Fourier Spherical Convolutional Networks

Mila, University of Montreal

Implicit planner GNNs for continuous control

Dr Alessandro Barp

University of Cambridge

Geometry of HMC and Geometric Integration for Sampling and Optimization

Imperial College London

Geometric Learning on Shapes and Distributions with Optimal Transport

Technion - Israel Institute of Technology

Self-supervised non-rigid correspondence by geodesic distortion minimization using the deformation field

Prof Søren Hauberg

Technical University of Denmark

Learning Latent Geometries

Technical University of Munich

Investigating Differentiable Graph Module

Dr Haggai Maron

Nvidia

Improved expressive power for message-passing networks via subgraph aggregation

Dr Challenger Mishra

University of Cambridge, University of Oxford

Characterising Universes in String Theory using Geometric Learning

Marie-Julie Rakotosaona

École Polytechnique

Learning to recover meshes from point clouds

Dr Petar Veličković

DeepMind

Navigating text adventures with algorithmic reasoners