This is an old revision of the document!
Video of the lectures will be available via the echo360 portal of the course
Topics | Reading | Assignments | |
---|---|---|---|
Week 1: August 25,27 | |||
Tuesday | Course introduction ( slides). | Sections 1.1 and 1.2 in the textbook | |
Thursday | Course introduction (continued). Linear models and the perceptron algorithm ( slides) | Chapters 1,3.1 in the textbook | |
Week 2: September 1,3 | |||
Tuesday | Linear models (continued). Short intro to python [ notes ] | Chapters 1,3.1 in the textbook | Assignment 1 is available. Due date: 9/17. |
Thursday | More Python; code for the perceptron. Linear regression ( slides) | Chapter 3.2 | |
Week 3: September 8,10 | |||
Tuesday | Linear regression (continued). Intro to latex | Chapter 3.2 | |
Thursday | Logistic regression ( slides) | Chapter 3.3 | |
Week 4: September 15,17 | |||
Tuesday | Overfitting ( slides) | Chapters 2.3,4.1 | |
Thursday | Regularization and model selection; cross validation ( slides) | Chapter 4.2, 4.2.2 | Assignment 2 is available. Due date: 10/2. |
Week 5: September 22,24 | |||
Tuesday | Support vector machines ( slides) | Chapter e-8 | |
Thursday | SVMs (continued) | Chapter e-8 | |
Week 6: September 29, October 1 | |||
Tuesday | Expressing SVMs in terms of error + regularization; unbalanced data ( slides). Here's code for displaying the decision boundary of a classifier. | Chapter e-8 | |
Thursday | Nonlinear SVMs: kernels ( slides) | Chapter e-8 | Assignment 3 is available. Due date: 10/16. |
Week 7: October 6,8 | |||
Tuesday | Kernels continued; model selection ( slides); a demo of model selection in scikit-learn. | Chapter e-8 | |
Thursday | Multi-class classification ( slides). And here's how to do it in scikit-learn. | ||
Week 8: October 13,15 | |||
Tuesday | Neural networks and the backpropagation algorithm ( slides) | Chapter e-7 | |
Thursday | Neural networks (continued) code for neural networks trained using backpropagation | Chapter e-7 | Assignment 4 is available. Due date: 10/30. |
Week 9: October 20,22 | |||
Tuesday | Neural networks (continued) | Chapter e-7 | |
Thursday | Deep networks ( slides) | Chapter e-7 | |
Week 10: October 27,29 | |||
Tuesday | Deep networks (continued) | Chapter e-7 | |
Thursday | Features and feature selection ( slides) and here is some code for feature selection. | Chapter e-9 | |
Week 11: November 3,5 | |||
Tuesday | Principal components analysis ( slides) | Chapter e-9 | Assignment 5 is available. Due date: 11/15. |
Thursday | Nearest neighbor methods ( slides) | Chapter e-6 | |
Week 12: November 10,12 | |||
Tuesday | Clustering ( slides) | Chapter 10 in introduction to statistical learning | |
Thursday | Clustering (cont); stability-based model selection for clustering ( slides) | A. Ben-Hur, A. Elisseeff and I. Guyon. A stability based method for discovering structure in clustered data. Pacific Symposium on Biocomputing, 2002. |