Machine Learning - WS 2011/2012


Roland Memisevic
Office hours: Mondays, 4pm - 5pm.
To contact instructor or tutors, send mail to

Time and Place

Lectures: Monday 2pm-4pm, Robert-Mayer-Str. 11-15, SR9
Tutorials: Wednesday 2pm-4pm, Robert-Mayer-Str. 11-15, SR307


Some knowledge of calculus, linear algebra and statistics.


The official course textbook is: Christopher M. Bishop: Pattern Recognition and Machine Learning

Other useful books include
David MacKay: Information Theory, Inference, and Learning Algorithms
Hastie, Tibshirani, Friedman: The Elements of Statistical Learning
Duda and Hart: Pattern Classification


Written or oral exam.
Taking part in the tutorials is not binding but highly recommended.
You can improve your exam grade by up to one mark when successfully taking part in the exercises.

Final Exam

Place: The usual Wednesday tutorial Room (SR307).
Start: 2pm. Duration: 120 minutes.

Python resources useful for some of the exercises

General Python:

official tutorial

Packages useful for ML:


Provisional Schedule

Date Topic Readings Materials
Oct. 17 Lecture Introduction Book, Chapter 1 notes
Oct. 19 Tutorial Background on probabilities Chapters 1.2, (2.1-2.3) notes
Oct. 24 Lecture Linear regression Chapter 3 notes
Oct. 26 Tutorial Numerical data processing in Python notes
Oct. 31 Lecture Linear classification Chapter 4 notes
assignment 1
Nov. 2 Tutorial Questions about Assignment 1 and lecture
Nov. 14 Lecture Non-linear models and back-propagation Chapters 5.1-5.3, 5.5 notes
Nov. 16 Tutorial Discussion of Assignment 1 solutions to programming exercises
Nov. 21 Lecture Clustering and the EM algorithm Chapters 2.3.9, 9.1-9.4 notes
assignment 2
Nov. 23 Tutorial Questions about Assignment 2 and lecture
Nov. 28 Lecture Continuous latent variables and PCA Chapters 12.1, 12.2, 12.4 notes
Nov. 30 Tutorial Discussion of Assignment 2 solutions to programming exercises
Dec. 5 Lecture Sequences and Hidden Markov Models Chapters 13.1, 13.2
assignment 3
Dec. 7 Tutorial Questions about Assignment 1 and lecture
Dec. 19 Lecture Graphical models Chapter 8 notes
Dec. 21 Tutorial Discussion of Assignment 3 solutions to programming exercises
Jan. 9 Lecture Approximate inference and sampling Chapters 10.1, 11.1-11.3 notes
assignment 4
Jan. 11 Tutorial Questions about Assignment 4 and lecture
Jan. 16 Lecture Overfitting, capacity control, Bayesian reasoning Chapters 2.1, 2.2, 2.3, 3.3, 3.4, 3.5 notes
Jan. 18 Tutorial Review
Jan. 23 Lecture Nonparametric methods, kernel methods,
support vector machines
Chapters 6, 7.1 notes
Jan. 25 Tutorial Discussion of Assignment 4 solutions to programming exercises
Jan. 30 Lecture Unsupervised feature learning and deep learning Useful (though very detailed) background reading includes:
"Learning Deep Architectures for AI", Y. Bengio, 2009
or google: "deep unsupervised learning"
Feb. 6 Lecture special lecture, not relevant for the exam
Feb. 8 Final exam Place: The usual Wednesday tutorial Room (SR307).
Start: 2pm. Duration: 120 minutes.

The information provided here may change at any time.

Roland Memisevic