Emtiyaz Khan

Machine Learning Researcher & Lecturer at EPFL (IC)
Home News Research Publications Teaching Blog Resume Contact
Emti's photo

Recent news

My talk at AI-Stats 2014


  • Goal 1: To design models to learn from data that are unreliable, noisy, high-dimensional, heterogenous, missing, and large.
    Topics : Bayesian Models, Probabilistic Graphical Models, and Latent Variable Models.
  • Goal 2: To design algorithms that are accurate, fast, scalable, and easy to use, all at the same time.
    Topics: Variational Inference, Stochastic non-convex optimization, Bayesian Optimization.

  List of current projects

  • Stochastic-approximation variational inference.
  • Variance reduction for stochastic variational inference.
  • Application of variational inference to deep learning.
  • Scalable inference for Gaussian process models.
  • Distributed optimization for variational inference.
  • UAVs doing Bayesian optimization to track humans.
  • Online collaborative prediction of vote outcomes.
  • Predicting success of crowfunded projects.
  • Personalization and breaking the filter bubble.
  • Home monitoring of patients.