CS 534
Machine Learning

Lecture Topics and Reading
  Spring 2007

Date

Topic

Recommended
Reading

Notes

April 2
Introduction, basic concepts
Chapter 1
Probability Theory Review: A.4 and Andrew Ng's review
PDF
April 4
Perceptron algorithm Section 5.1,2,4,5 PDF
April 6
Logistic regression class notes
PDF
April 9
Linear Discriminant Analysis (LDA)
Section 2.1-6 PDF
April 11
cont. LDA.


April 13
Off-the-Shelf Classifiers, Decision Trees (DT)
Pattern Classification Chapter 8 or
Macine Learning Chapter 3
PDF
April 16
cont. DT, Neural Networks (NNET)
Pattern Classification Chapter 6
PDF
April 18
cont. NNET,  Nearest Neighbor Classifiers (NNC)
Optional Reading: Machine Learning Chapter 8
PDF
April 20
cont. NNC, Support Vector Machines (SVMs)
Optional Reading
Pattern Classification Section 5.2;
C. J. C. Burges. A Tutorial on Support Vector Machines for Pattern Recognition. Knowledge Discovery and Data Mining, 2(2), 1998
PDF
April 23
cont. SVMs

PDF
April 25
Naive Bayes Classifiers
Optional Reading:
Generative and Discriminative Classifiers:Naive Bayes and Logistic Regression , chapter by Tom Mitchell for the 2nd edition of his Machine Learning book
PDF
April 27
Computational Learning Theory (COLT):
PAC Learning
Optional: Machine Learning Chapter 7
PDF
April 30
COLT: Inconsistent Hypotheses and VC-Dimension Bounds
Optional: Machine Learning
Chapter 7
PDF
May 2
cont. COLT, Model Selection and Regularization
Pattern Classification Chapter 9
Machine Learning Chapter 6 (6.1-6.6)
PDF
May 4
Midterm Review , cont. Model Selection and Regularization

May 7
Ensemble Learning: Bagging, Equivalence of Weak and Strong Learnability Pattern Classification Chapter 9 (9.4-5) PDF
May 9
AdaBoost, Review Final Project Pattern Classification Chapter 9 (9.4-5)
May 11
Midterm (Different Location: BEXL 412)

May 14

Functional Gradient Boosting
"Greedy Function Approximation: A Gradient Boosting Machine", Jerome Friedman (required Sections 1,2,3, 4.4)
PDF
May 16
cont. Gradient Boosting, Sequential Supervised Learning (SSL)


May 18
cont. SSLw/ Hidden Markov Models
Pattern Classification 3.10 (through 3.10.5) PDF
May 21
cont. SSL w/ HMMs


May 23
SSL w/ Perceptron
Supplemental Reading:
"Discriminative Training Methods for Hidden Markov Models: Theory and Experiments with Perceptron Algorithms"
Michael Collins, EMNLP 2002
PDF
May 25
SSL w/ Conditional Random Fields
Supplemental Reading:
"Conditional Random Fields: An Introduction", Hanna Walsh,
Technical Report MS-CIS-04-21. Dept. of Computer and Information Science, University of Pennsylvania, 2004.
PDF
May 28
Memorial Day: No Class


May 30
Unsupervised Learning: Hierarchical Agglomerative Clustering
Pattern Classification 10.9
PDF
June 1
K-means clustering, Probabilistic Clustering
Optional Reading:
Pattern Classification 10.4.1-10.4.3

June 4
Expectation Maximization, Dimensionality Reduction

PDF
June 6
Pricipal Component Analysis, Semi-Supervised Learning

PDF
June 8

Finish Semi-Supervised Learning, Review, Q/A

PDF
June 14
Final Examination: 9:30a-11:30a


.