Random Matrices Physics and Machine Learning


The CFM-ENS chair “Modèles et Sciences des Données” organize a workshop on Data Sciences on November 10th, from 14:00 to 18:00 in the conference room of the ENS Center for data sciences, 45 rue d’Ulm (Stair B 3rd floor).

When

Wednesday, November 10th 2021, 14:00-18:00.

Where

Conference room of the ENS Center for data sciences, 45 rue d’Ulm (Stair B 3rd floor).

Schedule

  • 14h00-14h45: Romain Couillet (Grenoble-Alps University) Random matrices could steer the dangerous path taken by AI but even that is pointless
  • 14h45-15h20: Marc Potters (Capital Fund Management) Optimal block size in cross-validation: an example form covariance matrix estimation
  • 15h20-15h45: Maria Refinetti - Stephane d’Ascoli (Laboratoire de Physique and CDS ENS) Reconciling double descent with older paradigms
  • 15h45 16h15: Coffee break
  • 16h15-16h50: Sylvain Gigan (LKB ENS and Sorbonne University) Light in complex media: random matrices for optical imaging and computing
  • 16h50-17h15: Valentina Ros (LPTMS Paris-Saclay University) Counting critical points of random landscapes: Kac- Rice formulas and random matrix theory
  • 17h15-17h40: Rudy Morel (Departement d’Informatique and CDS ENS) Time-scale stationary processes for finance and turbulence