Past edition: Statlearn’14, April 7-8, 2014

Program of Statlearn’14 and links to slides and videos of the talks

  • Monday April, 7
    • 9h30 – 10h : Coffee
    • 10h – 12h : Sparsity in high-dimensional problems. Slides and videos
      • Shakir Mohamed (University of London): Bayesian and L1 Approaches for Sparse Unsupervised Learning
      • François Caron (University of Oxford): Probabilistic Low-Rank Matrix Completion with Adaptive Spectral Regularization Algorithms
    • 14h-17h30 : Learning and applications in life science. Slides and videos
      • Stéphane Mallat (Ecole Polytechnique): High-Dimensional Classification with Deep Scattering Networks
      • Florence Forbes (INRIA Rhône-Alpes): Variational approach for the analysis of fMRI data and hemodynamically informed parcellation of the brain
      • Pierre Neuvial (Université d’Evry, Stat & Génome): Performance evaluation of DNA copy number segmentation methods
  • Tuesday April, 8
    • 9h30 – 12h30 : Beyond passive learning. Slides and videos
      • Vincent Lemaire (Orange Labs): Real World Issues in Supervised Classification for data stream
      • Aurélien Garivier (Université de Toulouse): Optimal Discovery with Probabilistic Expert Advice: Finite Time Analysis and Macroscopic Optimality
      • Sébastien Bubeck (Princeton University): On the influence of the seed graph in the preferential attachment model
    • 14h-17h30 : Future problems in statistical learning. Slides and videos
      • Eric Moulines (TelecomParisTech): On stochastic proximal gradient algorithm
      • Alain Celisse (Université Lille 1 & INRIA): Detecting changes in the distribution with kernels
      • Julien Mairal (INRIA Rhône-Alpes) : Incremental and Stochastic Majorization-Minimization Algorithms for Large-Scale Machine Learning