Fall, 2005

** Students should consult this page at least once a week
for important information. **

** Lectures: ** Thursdays 12-2pm, MS 2173

** Instructor: ** Toniann Pitassi, email: toni@cs

Office: Sandford Fleming 2305A, 978-3695

Office Hours: Thursday 3-4pm

** Teaching assistant: ** Mark Braverman, email: mbraverm@cs

**
LECTURE NOTES
**

**
**

- Lecture 1
- Lecture 2
- Lecture 3
- Lecture 4
- Lecture 5
- Lecture 6
- Lecture 7
- Lecture 8
- Lecture 9
- Lecture 10
- Lecture 11
- Lecture 12

** SCRIBE HELP **

** HOMEWORK **

** PAPERS AND OTHER READINGS **

- Learning Quickly when Irrelevant attributes abound: A New Linear Threshold Algorithm Paper by Nick Littlestone introducing online learning model and the Winnow algorithm.
- Handout by Avrim Blum on Tail Inequalities
- The Weighted Majority Algorithm Paper by Littlestone and Warmuth introducing the weighted majority algorithm.
- On-Line Algorithms in Machine Learning Excellent survey article.
- Chapter on PAC learning model, and decision-theoretic generalizations with applications to neural nets Survey of the PAC model and sample-complexity results.
- A Theory of the Learnable Original paper by Leslie Valiant introducing the PAC model.
- A decision-theoretic generalization of on-line learning and an application to boosting Original paper introducing Adaboost.
- Boosting the margin: A new explanation for the effectiveness of voting methods Great paper giving margins analysis of Adaboost.
- The Boosting approach to Machine Learning: On Overview Survey article on boosting by Rob Schapire.
- Boosting and hard-core distributions by Klivans and Servedio
- Survey article on the Multiplicative Weights Update Method
- Hard core distributions from somewhat hard problems by Russell Impagliazzo
- New Results on Hardness of Proper Learning and Beyond `
- Weakly learning DNF and Characterizing Statistical Query learning using Fourier Analysis
- Learning DNF in time exp(n^{1/3}) Best known algorithm for learning DNF.
- Learning thresholds and intersections of halfspaces
- Vapnik paper on Support Vector machines
- Support Vector Networks Survey article on support vector machines.
- Some PAC Bayesian Theorems Excellent paper by McAllester.
- A Proof of MacAllester's PAC Bayesian Theorem.

** RELATED COURSE **