Spring 2006
| Machine Learning home page | Syllabus | Schedule/Readings | Project Assignments | Resources |
| Date | Topics | Assigned Readings |
| Thurs. 1/12 |
Course Introduction Intro. to Learning |
Ch. 1 |
| Tues. 1/17 |
Reinforcement Learning Elements of RL problem |
Ch. 16 |
| Thurs. 1/19 |
Reinforcement Learning (con't.) K-armed bandit State-action pairs Defining learning rate Exploration vs. exploitation |
Ch. 16 (con't.) Class handout (Ch. 21 from Russell and Norvig) |
| Tues. 1/24 |
Reinforcement Learning (con't.) Q-learning Sarsa Sarsa(lambda) |
Ch. 16 (con't.) Project 1 assigned; due Feb. 12 |
| Thurs. 1/26 |
Player/Stage tutorial (to be used in Project 1; taught by Michael Bailey) |
Player/Stage
Getting Started Guide (UTK-specific)
Player/Stage Documentation (Public-domain website) |
| Tues. 1/31 |
Reinforcement Learning (con't.) Case Studies Practical implementation issues |
"Getting Reinforcement Learning to Work on Real Robots",
by Smart and Kaelbling, Artificial Intelligence, 55 (2-3), 311-365, 1991.
"Learning to Coordinate Behaviors", by Maes and Brooks, Proc. of 8th Nat'l. Conf. on Artificial Intelligence (AAAI-90), AAAI Press/MIT Press, pgs. 796-802, 1990. Improving Elevator Performance Using Reinforcement Learning, In Touretzky, et al (eds.), Advances in Neural Information Processing Systems: Proc. of the 1995 Conference, MIT Press, pgs. 1017-1023, 1996. |
| Thurs. 2/2 |
Neural Networks Introduction |
Ch. 11
"The Basic Ideas in Neural Networks",
by Rumelhart et al., Communications of the ACM, 37(3): 87-92, 1994.
Applet links (just to play around with): |
| Tues. 2/7 |
Neural Networks (con't.) Perceptrons |
Ch. 11 (con't.)
Applet link (just to play around with): |
| Thurs. 2/9 |
Neural Networks (con't.) Multi-layer feedforward NNets Back Propagation |
Ch. 11 (con't.) Project 2 assigned; due Mar. 7 |
| Tues. 2/14 |
Neural Networks (con't.) K-fold cross-validation Case studies: Face Recognition Design and performance issues |
Ch. 14.2.1 (pg. 331) [See also top of today's Mitchell handout]
Class handout (face recognition case study from Mitchell) "Analysis of Hidden Units in a Layered Network Trained to Classify Sonar Targets", by Gorman and Sejnowski, Neural Networks, Vol. 1, pgs. 75-89, 1988. Ch. 11 (con't.) |
| Thurs. 2/16 |
In-class design competition #1 (in teams): Using neural networks or reinforcement learning for the Paddle Ball task (competing for extra credit points) |
Lecture notes on neural nets |
| Tues. 2/21 |
Review of in-class design exercise Review of Project #1 Introduction to Genetic Algorithms |
Feedback: In Class Design #1 Exercise Class handout (Ch. 9 from Mitchell on Genetic Algorithms) |
| Thurs. 2/23 |
Genetic Algorithms (con't.) Representation of hypotheses Genetic operators Fitness selection methods Examples: DNF satisfiability Traveling Salesman Problem |
Applet link (just to play around with): Traveling Salesman Problem Genetic Algorithm applet (go towards bottom of page) |
| Tues. 2/28 |
Genetic Algorithms (con't.) Genetic Programming Block-stacking example |
|
| Thurs. 3/2 |
Genetic Programming (con't.) Boolean 11-multiplexer example Design details Artificial ant example |
|
| Tues. 3/7 |
Genetic Programming (con't.) Symbolic regression example |
Project 3 assigned; Part I due at 23:59:59 on Wed. Mar. 15 Part II due at 08:00 on Thurs. April 6
Class handout (GP examples from Koza text)
Class handout (GP details -- parts of Ch. 6 of Koza text)
Videos: evolutionary computation
(for designing robots) |
| Thurs. 3/9 |
Introduction to density estimation Parametric Learning |
Ch. 1.2.4, Ch. 4 |
| Tues. 3/14 |
In-class design competition #2 (in teams) Using genetic programming for the pole-balancing task (competing for extra credit points) | |
| Thurs. 3/16 |
No class; instructor on travel (Work on those projects!!) |
|
| Tues. 3/21 | Spring Break; no class | |
| Thurs. 3/23 | Spring Break; no class | |
| Tues. 3/28 | Multivariate parametric learning | Ch. 5.1-5.4 |
| Thurs. 3/30 | Multivariate classification | Ch. 5.5-5.7 |
| Tues. 4/4 |
Dimensionality reduction
Subset Selection Principal Components Analysis (PCA) |
Ch. 6.1-6.3
Applet link (just to play around with):
on PCA
|
| Thurs. 4/6 |
(Briefly) Factor Analysis K-Means Clustering |
Project 4 assigned; due April 27 Ch. 6.4 Ch. 7.3 |
| Tues. 4/11 |
Fuzzy C-Means Clustering Case Study: Gesture Recognition Hierarchical Clustering |
Handout: X. Li paper (unpublished) on gesture recognition using Fuzzy C-Means clustering Ch. 7.7 |
| Thurs. 4/13 |
Self-Organizing Maps (Overview) |
Class handout (SOMs applied to EEGs): "Self-Organizing Map in Recognition of Topographic Patterns of EEG Spectra", by Joutsiniemi, Kaski, and Larsen, IEEE Transactions on Biomedical Engineering, Vol. 42, No. 11, pgs. 1062-1068 Ch. 12.2.3 Applet link (just to play around with): on SOMs |
| Tues. 4/18 |
Lazy Learning (Overview) (i.e., Instance-based and case-based learning) Tutorial slides on Instance Based learning |
Ch. 8 |
| Thurs. 4/20 |
Anomaly Detection (Overview) (applied to biosurveillance case study) Tutorial slides anomaly detection, applied to biosurveillance |
|
| Tues. 4/25 |
Introduction/Overview of Project 5 Combining Multiple Learners (Overview) |
Project 5 (Poster) assigned; to be presented
in Poster Session during final exam period (May 4) Hints on how to prepare a poster Ch. 15 |
| Thurs. 4/27 |
Combining Multiple Learners (con't.) "Final Quiz" (doesn't count! just for extra credit) |
|
| Thurs. 5/4, 7:15PM - 9:15PM | Poster Session (during scheduled final exam time period) |