"Bayesian Statistics & Machine Learning" Reading Group at Northwestern Statistics

This page serves as a repository for resources of the 2018-2019 reading group "Bayesian Statistics & Machine Learning" at Department of Statistics at Northwestern University. In Fall 2018, this is set to be held weekly/bi-weekly (flexible) on Tuesday 3:30pm-5pm.

We mainly follow the course materials of "Bayesian Models for Machine Learning" and Advanced Probabilistic Machine Learning by Prof. John Paisley at Columbia. The past course notes can be accessed here (for BMML) and here (for APML).

Several textbooks can also be used for reference, including (but not limited to):

1. Pattern Recognition and Machine Learning by Christopher M. Bishop
2. Information Theory, Inference, and Learning Algorithms by David J.C. MacKay
3. Machine Learning: A Bayesian and Optimization Perspective by Sergios Theodoridis

To supplement the theory content in above materials, we will also learn Python with TensorFlow (a computational library by Google, mainly for deep learning but is also a general purpose optimizer) with TensorFlow Probability (a library for probabilistic reasoning and statistical analysis in TensorFlow). The example code will be maintained in this GitHub repository.

Good references of Python (Scikit-Learn) and TensorFlow for machine learning, including (but not limited to):

1. Hands-On Machine Learning with Scikit-Learn & TensorFlow by Aurelien Geron

Schedule (tentative):

• 10/16 (Tue, 3:30-5pm, Tim):

  • Theory: Probability review, Bayes rule, conjugate priors (Lecture 1 of BMML Notes)
  • Practical: Introduction to Python

• 11/6 (Tue, 3:30-5pm, Tim):

  • Theory: Bayesian linear regression, Bayes classifiers, predictive distributions (Lecture 2 of BMML Notes)
  • Practical: Introduction to TensorFlow

• 11/13 (Tue, 3:30-5pm):

  • Theory:
  • Practical: