Teaching - http://www.cs.ualberta.ca/~dlizotte/teaching/
Introduction to Machine Learning - Winter 2008
Guest lecture: Introduction to Gaussian Processes.Introduction to Machine Learning - Spring 2007 - Reykjavík University
In the age of Google and Netflix, enabling computers to learn from data has become a prerequisite for solving many interesting problems. This course will cover the foundations of machine learning along with basic algorithms used in supervised learning, unsupervised learning, and reinforcement learning. Topics covered will include decision trees, support vector machines, and k-nearest-neighbours for classification, as well as k-means clustering for unsupervised learning. Reinforcement learning topics will include offline and online algorithms for agents learning to behave optimally in uncertain environments. Students will have the opportunity to gain hands-on experience with techniques in each of these areas.
Topics and Lecture Notes
These are drawn from three main sources: Andrew Moore's excellent series of tutorials in machine learning, Russ Greiner's teaching materials, and Rich Sutton's slides on reinforcement learning.
- Introduction
- Alpaydin Chapter 1
- Decision Trees
- Alpaydin Chapter 9.1-9.3
- AIxploratorium
- PAC Learning
- Alpaydin Chapter 2.3
- VC Dimension
- Alpaydin Chapter 2.2
- Wikipedia entry on VC Dimension
- Overfitting and Cross Validation
- Alpaydin Chapter 14.1 - 14.2
- Probabilities, Bayes Classifiers - Part 1
- Probabilities, Bayes Classifiers - Part 2
- Alpaydin Appendix A.1, A.2.1 - A.2.4
- Wikipedia entry on Naive Bayes
- Alpaydin Chapter 3.7 - but this is more complicated than we need. If you are interested in general Bayes nets, look here.
- Support Vector Machines - Part 1
- Support Vector Machines - Part 2
- Alpaydin Chapter 10.9
- Reinforcement Learning - Part 1
- Alpaydin Chapter 16
- Sutton and Barto Chapters 1, 2, 3
- Reinforcement Learning - Part 2: TD
- Alpaydin Chapter 16.5
- Sutton and Barto Chapter 6
- Exam Review
Assignments
Assignments are due at the beginning of class on the date specified.
Late assignments will not be accepted.
- Assignment 1 - Due on April 30th
- Assignment 2 - Due on May 7th
- Assignment 3 - Due on May 14th at noon via e-mail
- RUGridWorldDemo Reinforcement Learning Envrionment
- Assignment 3 Part 2 sample solution - PDF
- Assignment 3 Part 2 sample solution - Excel
| Required text: | Supplemental text: |
|---|---|
|
Introduction to Machine Learning Ethem Alpaydin The MIT Press ISBN-10: 0-262-01211-1 ISBN-13: 978-0-262-01211-9 |
Reinforcement Learning: An Introduction Richard S. Sutton and Andrew G. Barto The MIT Press ISBN-10: 0-262-19398-1 ISBN-13: 978-0-262-19398-6 FULL TEXT AVAILABLE ONLINE |
