November 6 - 8, 2024
November 6 - 8, 2024
We aim for ample discussion time within the workshop. The workshop program will start on Wednesday morning and end on Friday by lunch to allow convenient same-day travel. If you are only partially available please contact one of the organizers. The mini-courses will take place on Monday and Tuesday.
Tentative program:
1. Session: Laplacians and graphical models in extremes
Talk 1: Ignacio Echave-Sustaeta Rodríguez (TU Eindhoven): Cayley–Menger matrices, total correlation and extremal conditional independence
Talk 2: Jane Coons (MPI-CBG Dresden): Algebraic Statistics for Huesler-Reiss Graphical Models
2. Session: Graph signal processing & Compositional data analysis
Talk 1: Manuel Hentschel (University of Geneva): Rank-Deficient Covariance Structures in Extreme Value Theory and Compositional Data Analysis
Talk 2: Elvin Isufi (Delft University of Technology): Relational Learning via Covariance Information
Talk 3: Gecia Bravo-Hermsdorff (University of Edinburgh): Stochastic graph reduction via combinatorial invariants
3. Session: Theory of graph Laplacians
Talk 1: Karel Devriendt (University of Oxford): Geometry of Laplacian matrices: effective resistances, simplices and new invariants
Talk 2: Angeles Carmona (Universitat Politècnica de Catalunya): Equilibrium Measures for Laplacian matrices in Random Walks
Talk 3: Aida Abiad (TU Eindhoven): A spectral approach to Kemeny’s constant
4. Session: Graphical extremes I
Talk 1: Kirstin Strokorb (Cardiff University): Graphical models for infinite measures with applications to extremes
Talk 2: Sebastian Engelke (University of Geneva): Graphical models for Lévy processes
Talk 3: Johan Segers (KU Leuven and UCLouvain): X-vines, Hüsler—Reiss distributions, and partial correlations
5. Session: Graph signal processing
Talk 1: Fabio Vitale (CENTAI Institute (Italy)): Fast and Effective GNN Training through Sequences of Random Path Graphs
Talk 2: Michal Valko (Meta / Inria / ENS): Predictive and non-constrastive self-supervised representation learning for graphs
6. Session: Graphical extremes II
Talk 1: Nicola Gnecco (UCL): Directed graphical models and structural causal models in multivariate extremes from threshold exceedances
Talk 2: Marco Oesting (University of Stuttgart): Scalable Algorithms for High-Dimensional Extremes
Talk 3: Chen Zhou (Erasmus University Rotterdam): Graphical lasso for extremes
7. Session: Algebraic Statistics
Talk 1: Pratik Misra (TU Munich): Graph Laplacians and Gaussian models of ML degree 1
Talk 2: Carlos Améndola (TU Berlin): Likelihood Geometry of Hüsler-Reiss graphical models
Conference dinner: The conference dinner will take place in Restaurant Stories, about 30 minutes on foot or 6 by bus from the central train station.
Social program: We are organizing a tour of the PSV Eindhoven football stadium on Thursday afternoon. Please send an email to Ignacio Echave-Sustaeta Rodríguez if you would like to participate.