Bayesian Machine Learning and Information Processing (5SSD0)

academic year 2020/21

The 2020/21 course “Bayesian Machine Learning and Information Processing” will start in November 2020 (Q2).

Course summary

This course covers the fundamentals of a Bayesian (i.e., probabilistic) approach to machine learning and information processing systems. The Bayesian approach allows for a unified and consistent treatment of many model-based machine learning techniques. We focus on Linear Gaussian systems and will discuss many useful models and applications, including common regression and classification methods, Gaussian mixture models, hidden Markov models and Kalman filters. We will discuss important algorithms for parameter estimation in these models including the Expectation-Maximization (EM) algorithm and Variational Bayes (VB). The Bayesian method also provides tools for comparing the performance of different information processing systems by means of estimating the ``Bayesian evidence’’ for each model. We will discuss several methods for approximating Bayesian evidence. Next, we will discuss intelligent agents that learn purposeful behavior from interactions with their environment. These agents are used for applications such as self-driving cars or interactive design of virtual and augmented realities. Indeed, in this course we relate synthetic Bayesian intelligent agents to natural intelligent agents such as the brain. You will be challenged to code Bayesian machine learning algorithms yourself and apply them to practical information processing problems.

Course goals

This course provides an introduction to Bayesian machine learning and information processing systems. The Bayesian approach affords a unified and consistent treatment of many useful information processing systems.

News and Announcements

Instructors

Materials

In principle, you can download all needed materials from the links below.

Books

Please download the following books/resources:

  1. Christopher M. Bishop (2006), Pattern Recognition and Machine Learning. You can also buy a hardcopy, e.g. at bol.com.
  2. Ariel Caticha (2012), Entropic Inference and the Foundations of Physics.
  3. Bert de Vries et al. (2020), PDF bundle of lecture notes for lessons B0 through B12 (Ed. Q3-2019/20).
    • The lecture notes may change a bit during the course, e.g., to process comments by students. A final PDF version will be posted after the last lecture.
  4. Wouter Kouw (2020), Julia and Jupyter Install Guide.
    • Use this guide if you need help to install Julia and Jupyter, so that you can open and run the course notebooks on your own machine.
    • You can test your installation by running the notebook called “Probabilistic-Programming-0.ipynb”, which can be downloaded from github (under lessons/notebooks/probprog). Here is a video with step-by-step instructions on opening course notebooks.

Lecture notes and videos

The source files for the lecture notes are accessible on github. If you want to download them, click the green Code button and then Download ZIP. The theory lectures are under lessons/notebooks and the programming notebooks are under lessons/notebooks/probprog. Note that you don’t have to download them, you can view all lecture notes online through the links below:

Date lesson materials
video guides lecture notes live class recordings
11-Nov-2020 B0: Course Syllabus and Administrative Issues
B1: Machine Learning Overview
B1 B0, B1 B0
13-Nov-2020 B2: Probability Theory Review B2.1, B2.2 B2 B2
18-Nov-2020 B3: Bayesian Machine Learning B3.1, B3.2 B3 B3
20-Nov-2020 W1: Probabilistic Programming 1 - Intro Bayesian ML W1.1, W1.2, W1.3 W1 W1
25-Nov-2020 B4: Factor Graphs and the Sum-Product Algorithm B4 B4 B4
27-Nov-2020 B5: Continuous Data and the Gaussian Distribution B5.1, B5.2, B5.3 B5 B5
02-Dec-2020 B6: Discrete Data and the Multinomial Distribution B6 B6 B6, review B1-B6
04-Dec-2020 W2: ProbProg 2 - MP & Analytical Bayesian Solutions W2.1, W2.2, W2.3 W2 W2
09-Dec-2020 B7: Regression B7 B7 B7
11-Dec-2020 B8: Generative Classification
B9: Discriminative Classification
B8, B9 B8, B9 B8-9
16-Dec-2020 W3: ProbProg 3 - Regression and Classification W3.1, W3.2 W3 W3
18-Dec-2020 B10: Latent Variable Models and Variational Bayes B10 B10 B10
break
06-Jan-2021 B11: Dynamic Models B11 B11 B11
08-Jan-2021 B12: Intelligent Agents and Active Inference B12 B12 B12
13-Jan-2021 W4: ProbProg 4: Latent Variable and Dynamic Models W4.1, W4.2, W4.3 W4 PP4
15-Jan-2021 M1: Bonus Lecture: What is Life? M1 M1

Q&A

Q&A for each lesson can be accessed at the Piazza course site.

Exercises

In preparation for the exam, we recommend that you work through the following exercises to test your understanding of the materials:

Please feel free to consult the following matrix and Gaussian cheat sheets (by Sam Roweis) when doing the exercises.

Exam Guide

Each year there will be two exam opportunities. Check the official TUE course site for exam schedules. In the Q2-2020 course, your performance will be assessed by a WRITTEN EXAMINATION, which (very likely) will be offered both online (with proctoring software) and offline (on campus, if the situation allows it).

You cannot bring notes or books to the exam. All needed formulas are supplied at the exam sheet.

Instructors