COMP61011 : Foundations of Machine Learning



Lab sessions


Additional resources

Maintained by G.Brown
The material for the course can be found entirely in the course textbook.

The course is taught on WEDNESDAYS.

The lab sessions are 9am-12pm, in Kilburn Building.
You are assigned to a lab group dependent on your FAMILY NAME.
If your family name starts with a letter A-R, you are in room 2.25, upstairs.
If your family name starts with a letter S-Z, you are in room 1.8, on the middle floor.
The lectures are 1pm-5pm, in the Schuster Building
Weeks 1-3 are in the Bragg Theatre
Weeks 4-5 are in the Blackett Theatre

Week 1: Supervised learning, linear models, gradient descent, experimental methods.
Week 2: Support Vector Machines, Nearest Neighbour methods.
Week 3: Decision Trees, ROC analysis.
Week 4: Probabilistic models, Bayes Theorem.
Week 5: Ensemble Methods, Feature Selection.
Week 6: Independent project work

The structure of each week's session will be approximately:
0900-1200: Lab and Discussion of previous week's material, with demonstrators present.
1200-1300: Lunch
1300-1330: Unassessed multiple choice quiz on the previous week.
1330-1700: New lecture material.
The course text (provided in class) should be your main reference material. Associated with each week is a reading assignment. The unassessed MCQ will test this material, as well as the material from the previous week.

Week Taught in-classHomework Slides
1 (Wed 25th Sep) Ch 1+2Ch 3 Linear models
Part 1: PPT, PDF
Part 2: PPT, PDF
2 (Wed 2nd Oct) Ch 4Ch 5 Support Vector Machines
Part 1: PPT, PDF
Part 2: PPT, PDF
3 (Wed 9th Oct) Ch 6Ch 7 Intro to Neural Nets
Part 1 (PPTX), Part 2 (PPTX)
Decision Trees
4 (Wed 16th Oct) Ch 8Project (feedback opportunity in lab) Probabilistic Models
Part 1: PPT, PDF
Part 2: PPT, PDF

Now see also the Bayes Net tool developed by one of my former students!
5 (Wed 23rd Oct) Ch 9+10Project Ensemble Methods: PPT, PDF
Feature Selection PPT, PDF

6 (Wed 30th Oct) - Project deadline.
1st Nov (FRIDAY) 4pm.