Schedule of Lectures + Labs
The best way to learn something is to hear it many times, explained by different people. For every lecture/topic we cover, there are resources provided which explain the same thing, but in a subtly different way - some more mathematical, some more concept-oriented, some in more or less detail than others. It is highly advisable that you do not just rely on the lecture notes handed out in class to learn these topics. Your understanding will be greatly enhanced by additional reading.Printouts of all core material slides/notes will be provided in-class.
| Topics | Core Material | Supporting material | |
|---|---|---|---|
Tuesday 29th September | |||
| 09.15-09.30 | Welcome. |
Why Study Machine Learning? Slides | |
| 09.30-10.30 | Intro to Machine Learning. | Slides / Notes |
The Discipline of Machine Learning. |
| 11.00-12.00 | Support Vector Machines | Slides | Overview of SVMs - great survey material. |
| 13.00-14.30 | Lab : Orientation + Matlab. | Lab sheet 0 (Matlab) Matlab Tutorial |
Statistical Tests for ML Algorithms |
| 14.30-17.00 | Lab : Support Vector Machines | Lab sheet 1 (SVM) | A Practical Guide to SVM Classification |
Tuesday 6th October | |||
| 09.30-10.30 | Decision Trees | Slides / Notes |
www.DecisionTrees.net - fantastic resource. Information Gain Tutorial by Andrew Moore. |
| 11.00-12.00 | Feature Selection | Slides |
Lecture notes by Samy Bengio (Google) Intro to Feature Selection by Guyon+Elisseeff VideoLecture by Isabelle Guyon Fast Feature Selection with Conditional Mutual Info Recursive Feature Elimination with SVMs |
| 13.30-17.00 | Lab: Feature Selection. Marking of SVM lab | Lab sheet 2 (FS) | |
Tuesday 13th October | |||
| 09.30-10.30 | Probabilistic classifiers | Slides |
Basics of Probability video lecture. Bayes Theorem applet |
| 11.00-11.15 | Introduction to Project Work | ||
| 11.15-12.15 | Probabilistic Inference and Learning | Slides |
In an excellent book by David Mackay, Chapter 2 provides an introduction to
probabilistic inference. . The full book is here (do
not out the whole thing please!). Notes on Naive Bayes, extract from Barber's book |
| 13.30-17.00 | Lab : Begin Project Work, Marking of Feature Selection Lab. |
Project website | |
Tuesday 20th October | |||
| 09.30-10.30 | Clustering: Mixture models and the EM-algorithm | Slides |
Demo of Gaussian Mixture Models |
| 11.00-12.00 | Dimensionality reduction and PCA | Slides |
Notes on PCA, extract from Barber's book |
| 13.30-15.00 | Lab : Clustering | Lab sheet 3 (Clustering) | |
| 15.00-16.00 | Multiobjective Clustering (guest lecturer: Dr Joshua Knowles) | Slides |
|
| 16.00-17.00 | Lab : Clustering | Lab sheet 3 (Clustering) | |
Tuesday 27th October | |||
| 09.30-10.30 | Sequence Learning & Markov Chains | Slides |
|
| 11.00-12.00 | Hidden Markov Models | Slides |
Classic review article on HMMs by Rabiner |
| 13.30-17.00 | Lab : Sequence Learning, Marking session for Clustering lab. |
Lab sheet 4 (Sequence Learning) | |
Wednesday 4th November | |||
| 10.00-12.00 | Marking session for Sequence Learning lab. Submit Project Report to Student Support Office by Friday. |
||