Description

Students interested in Bayesian methods for machine learning can explore the state of the art in bayesian nonparametric methodologies and apply it to a real world problem. The objective is to do an overview survey of the state of the art and implement the different surveyed methods on a practical data set. This should be then benchmarked against parametric regression methods of your choice and a conclusion provided with the strengths and weaknesses of the Bayesian nonparametric methods

Objective

  • Delve into the literature of bayesian nonparametric
  • Apply the state of the art into a real world problem
  • Understand the weaknesses and strengths of BNP metods
  • Polish python skills

Data sets

  • https://www.kaggle.com/c/rossmann-store-sales/data
  • Private data set will be shared
  • https://www.kaggle.com/c/hack-reduce-dunnhumby-hackathon

Target audience

Ideal for students interested in Bayesian methods