Video of the lectures is available via the echo360 portal of the course
Topics | Reading | Assignments | |
---|---|---|---|
Week 2: Sept 2-6 | |||
Tuesday | Two simple linear models: the closest centroid algorithm and the perceptron algorithm ( slides) | Chapter 7 | assignment 1 is out |
Thursday | Evaluating and using ML classifiers( slides). And here's a demo of the process in PyML. | Chapter 2 | |
Week 3: Sept 9-13 | |||
Tuesday | Overview of Latex. Go over the code for the perceptron classifier. | Chapter 2,7 | |
Thursday | Classifier evaluation (continued) | Chapter 2 | |
Week 4: Sept 16-20 | |||
Tuesday | Linear regression ( slides). | Chapter 7 | |
Thursday | Linear regression - continued (6 slides were added to tuesday's batch). Here's code for ridge regression that you can try out in PyML. | Chapter 7 | Assignment 1 is due. Assignment 2 is out |
Week 5: Sept 23-27 | |||
Tuesday | Large margin classifiers: support vector machines ( slides). | Chapter 7 | |
Thursday | support vector machines (continued). | Chapter 7 |
Topics | Reading | Assignments | |
---|---|---|---|
Week 6: Sept 30 - Oct 4 | |||
Tuesday | SVMs and regularization; SVMs for unbalanced data ( slides) | A nice tutorial on SVMs: A user's guide to support vector machines. | |
Thursday | Extending SVMs to nonlinear classification ( slides). Here's a nice video that illustrates the idea. | Chapter 7 | Assignment 2 is due on Friday |
Week 7: Oct 7 - 11 | |||
Tuesday | Kernel classifiers: kernel versions of the perceptron and linear regression ( slides) and multi-class classification with binary classifiers (slides) | Chapter 7.5, Chapter 3 | Assignment 3 is out |
Thursday | Evaluating and using ML classifiers: model selection ( slides) | paper on Dataset selection | |
Week 8: Oct 14 - 18 | |||
Tuesday | More on kernel functions ( slides) | ||
Thursday | Kernel methods for protein-protein interactions ( slides) | A. Ben-Hur and W.S. Noble. Kernel methods for predicting protein-protein interactions. Bioinformatics 21(Suppl. 1): i38-i46, 2005. | |
Week 9: Oct 21 - 25 | |||
Tuesday | Distance based models and nearest neighbor classifiers ( slides) | Chapter 8 | Assignment 3 is due. Assignment 4 is out |
Thursday | Distance based clustering ( slides) | Chapter 8 | |
Week 10: Oct 28 - Nov 1 | |||
Tuesday | Probability theory, probabilistic models, and naive Bayes classification ( slides) | Chapter 9 | Assignment 4 is due. Assignment 5 is out |
Thursday | Continue discussion of naive Bayes. Obtaining probabilities from linear classifiers ( slides) | Chapter 7.4 |
Week 11: Nov 4 - Nov 8 | |||
---|---|---|---|
Tuesday | Logistic regression ( slides) | Chapter 9 | Assignment 4 is due. Assignment 5 is out |
Thursday | Features and feature selection ( slides) | Chapter 10 | Project proposal is due on friday |
Week 12: Nov 11 - Nov 15 | |||
Tuesday | Potential bias when using feature selection. Principal components analysis (PCA) ( slides) | Chapter 10 | |
Thursday | Decision trees ( slides) | Chapter 5 | |
Week 13: Nov 18 - Nov 22 | |||
Tuesday | Ensemble methods ( slides) | Chapter 11 | Assignment 5 is due. |
Thursday | An application of ML in bioinformatics: prediction of Calmodulin binding sites ( slides) | F.A. Minhas and A. Ben-Hur. Multiple instance learning of Calmodulin binding sites. Bioinformatics 28(18): i416-i422, 2012 |
Week 14: Dec 2 - Dec 6 | |||
---|---|---|---|
Tuesday | Neural networks ( slides) | ||
Thursday | Neural networks (cont); course summary ( slides) | ||
Week 15: Dec 9 - Dec 13 | |||
Tuesday | No class today | ||
Thursday | Student presentations | ||
Finals week: Dec 16 - Dec 20 | |||
Tuesday | Student presentations 5-8pm at CSB425 | ||
Thursday | Student presentations 5-8pm at CSB425 |