Schedule - Machine Learning Q3/2010

Week 1

Mon, Jan 25

Introduction to the course.

Crash course in statistics and probability theory.

Reading material:

• Bishop: 1.1-1.5, 2.1, 2.3.1 - 2.3.4, 2.4.1 - 2.4.2, 2.4.1 - 2.4.2.

Slides:

• Crash course in probability theory and statistics, part 1.

Thu, Jan 28

Crash course in statistics and probability theory, continued.

Reading material:

• Same as for Monday 25/1.

Slides:

• Crash course in probability theory and statistics, part 2

Week 2

Mon, Feb 1

Hidden Markov Models. Introduction, terminology and basic algorithms. Remember that the lecture starts at 1015!

Reading material:

• Bishop: 13.1 - 13.2
• Bishop: 8.1 - 8.2 (background reading)

Slides:

• Hidden Markov models - Terminology and Basic Algorithms

Thu, Feb 4

Hidden Markov Models. Implementation of basic algorithms.

Reading material:

• Bishop: 13.2.4 and 13.2.5 about implementing the forward- and backward-algorithms using scaled values
• You can also take a look at Appendix B - Floating Point Numbers from A. Tanenbaum, Structured Computer Organization, in order to se why we do not want numbers to become too small (i.e. entering the 'zone' of denormalized floats).

Slides:

• Hidden Markov models - Implementing the forward-, backward- and Viterbi-algorithms using log-space and scaling

Week 3

Mon, Feb 8

Hidden Markov Models. Traning and selecting model parameters.

Reading material:

• Bishop: 13.1 - 13.2, read 13.2.1, 13.2.2 and 13.2.5 in details, think about the equations as you read, you might want to read parts of chapter 9 as background. Honour students should read 9.4

Slides:

• Hidden Markov models - Selecting model parameters or "training"

Thu, Feb 11

Introduction to mandatory project 1 about Hidden Markov models (due Mar 1).

Reading material:

• Read the project description of mandatory project 1.
• For an alternative presetation of HMM methods and applications, you can take a look at this classic paper: Lawrence R. Rabiner. A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition. Proceeding of IEEE, 77:257-286, 1989.

Slides:

• Selecting the initial model parameters. Using HMMs for gene finding.
• Bits and pieces about project 1 (Gene findng).

Week 4

Mon, Feb 15

Q&A session about mandatory project 1.

Linear regression.

Reading Material:

• Bishop: 3.1, 3.4

Slides:

• Linear regression.

Thu, Feb 18

Linear regression and classification, continued.

Reading Material:

• Bishop: 4.1 - 4.1.5, 4.2 - 4.2.3 and 4.3.2

Slides:

• Linear Classification.

Week 5

Mon, Feb 22

No lecture. Time to work on project 1 ;-).

Thu, Feb 25

Clustering.

Reading material:

• Bishop: 9.1.

Slides:

• K-means clustering.

Week 6

Mon, Mar 1

Clustering, continued. Handin of mandatory project 1. Introduction to mandatory project 2.

Reading material:

• Clustering and Trees. Section 10.1-10.2 in "An Introduction to Bioinformatics Algorithms" by Neil C. Jones and Pavel A. Pevzner.
• The description of mandatory project 2.

Slides:

• Hiearchical clustering.

Thu, Mar 4

Neural networks.

Reading Material:

• Bishop: 5.1 - 5.3

Slides:

• Neural networks, part 1.

Week 7

Mon, Mar 8

Feedback on project 1. Information about the exam.

Slides:

• Feedback on project 1 and info about exam.

Thu, Mar 11

Neural networks.

Reading material:

• Bishop: 5.1 - 5.3

Slides:

• Neural Networks, part 2.