Deep Generative Modelling (Aalto University CS-E407509 2022)

Introduction

For each of the assigned chapters, you are required to prepare a couple of notes summarizing the chapter. There should be three sections covering

  1. Main points of the chapter
  2. Questions/unclear points
  3. Further comments/thoughts

There should be roughly three bullet points for each of them, without that being a hard constraint. For example, if you do not have any questions because everything is clear, then just focus more on the other two. The main focus of this task is not to spend too much time on this, or to submit overly long documents, but primarily for you to have a possibility to structure your thoughts while reading the chapter. We will then rely on these points to guide the discussion each week.

Content

Session 1

  • Chapter 10: Variational Inference

Read 10.1, 10.2.1, 10.2.3-5, 10.3.1, 10.3.2-3, 10.3.7-8, 10.5.1 Skim 10.2.2, 10.2.6, 10.3.4-6, 10.4, 10.5.2, 10.7 Skip 10.2.2, 10.2.7-8, 10.6

Session 2

  • Chapter 11: Markov Chain Inference
  • Chapter 12: Markov Chain Monte Carlo Inference

Session 3

  • Chapter 17: Bayesian Neural Networks

Session 4

  • Chapter 18: Gaussian Processes

Session 5

  • Chapter 22: Variational Autoencoders

Session 6

  • Chapter 23: Auto-regressive models

Session 7

  • Chapter 24: Normalizing flows

Session 8

  • Chapter 25: Energy-based models

Session 9

  • Chapter 26: Generative Adversarial Networks

Emacs 29.4 (Org mode 9.6.15)