Supplementary references are given for each lecture, either to research papers or to
the following textbooks:
Tom Mitchell, Machine Learning.
Stuart Russell and Peter Norvig, Artificial intelligence: a modern approach (second edition).
Michael Kearns and Umesh Vazirani, An introduction to computational learning theory.
Trevor Hastie, Robert Tibshirani, and Jerome Friedman, The elements of statistical learning.
Richard Duda, Peter Hart, and David Stork, Pattern classification.
Lecture 2: Linear separators: geometry; the perceptron algorithm [Jan 12]
Duda, Hart, and Stork -- 5.4, 5.5.
Lecture 3: Perceptron: experiments; voting perceptrons [Jan 17]
Slides: PPT PS
Lecture 4: Decision trees [Jan 19]
Slides: PPT PS
Russell and Norvig -- 18.3.
Mitchell -- 3.
Lecture 5: Probability review [Jan 24]
Mitchell -- 5.
Lecture 6: Large deviation bounds [Jan 26]
Kearns and Vazirani -- 9.
Lecture 7: Generalization [Jan 31]
Kearns and Vazirani -- 1,2,3.
Lecture 8: Generative models [Feb 2]
Duda, Hart, and Stork -- 2.
Lecture 9: Linear algebra background [Feb 7]
Lecture 10: The multivariate Gaussian [Feb 14]
Lecture 11: Generative models for classification [Feb 16]
Hastie, Tibshirani, and Friedman -- 4.
Lecture 12: Logistic regression [Feb 21]
Slides: PPT PS
Lecture 13: Unconstrained minimization [Feb 23]
Lecture 14: Kernels [Feb 28]
Slides: PPT PS
Lecture 15: Kernels [Mar 2]
Lecture 16: Support-vector machines [Mar 7]
A tutorial by Chris Burges.
Lecture 17: Support-vector machines [Mar 9]
Lecture 18: Weak learners; boosting [Mar 14]
Lecture 19: Multiclass classification [Mar 16]