COS424: Interacting with Data

Spring 2007


Information
Syllabus
Assignments

Syllabus

Note: this syllabus is tentative and subject to change.  Book readings refer to the books listed below, unless otherwise noted.  Some readings are available from the library's electronic reserves, for which you will need a userID and password that were distributed via email.  (Contact the course staff if you need them to be re-sent to you.)  You also can get these from blackboard using your OIT userID and password.

(RS)
Day Topic Reading Optional Reading
Tu, 2/6 Introduction scribe notes
Th, 2/8 Probability and statistics review scribe notes ; slides
CLASSIFICATION (RS)
Tu, 2/13 Introduction to classification and the K-nearest-neighbor algorithm scribe notes
Mitchell, pp.230-236
Th, 2/15 High dimensional space
 
scribe notes

Sections 9.5-9.6 of Coding and Information Theory (2nd ed, 1986) by R.W. Hamming

Decision trees Mitchell, Chapter 3
Tu, 2/20 Computational learning theory scribe notes

Mitchell, pp. 201-210

Th, 2/22 Boosting scribe notes

overview paper (sections 5,6,8 are optional)

boosting slides
face slide

Tu, 2/27 Support vector machines scribe notes

Sections 5.4-5.7 of The nature of statistical learning theory (1995) by V.N. Vapnik

Burges's tutorial on svm's
  CLUSTERING (DB)
Th, 3/1 K-means clustering scribe notes; slides

Hastie et al., Sections 14.1-14.3 (on e-reserve or blackboard)

Tu, 3/6 Agglomerative clustering scribe notes ; slides
  GRAPHICAL MODELS (DB)
Th, 3/8 Introduction to graphical models scribe notes Jordan's Statistical Science paper
Tu, 3/13 Naive Bayes classification case study slides ; scribe notes
Th, 3/15 Mixture models, latent variable models and EM scribe notes;
"Introduction to Graphical Models" 10.1, 11 (handout outside CS-204)
Tu, 3/27 Expectation-maximization scribe notes
REGRESSION (RS)
Th, 3/29
Tu, 4/3
Introduction to regression;
Linear regression
scribe notes 3/29
scribe notes 4/3
Hastie et al., pp. 41-45, 55-65, 115-120 (on e-reserve or blackboard);
Bishop, pp. 137-152, except sections 3.1.2 and 3.1.5
(there is some overlap between these readings that you can feel free to skim)
Th, 4/5 Logistic regression scribe notes
Generative and Discriminative Classifiers: Naive Bayes and Logistic Regression (new, unpublished chapter of the Mitchell book)
DIMENSIONALITY REDUCTION (DB)
Tu, 4/10
Th, 4/12
Principal components analysis scribe notes 4/10;
scribe notes 4/12;
Hastie et al., sections 14.5-14.6 (on e-reserve or blackboard)
Th, 4/19 Factor analysis scribe notes 4/19 ; MV Gaussian examples; Jordan Ch. 13-14 (outside of Dave's door)
ADVANCED TOPICS AND APPLICATIONS
Tu, 4/17 Maximum entropy modeling (RS) scribe notes
"A maximum entropy approach to species distribution modeling"
slides;
paper on modeling "bake-off"
Tu, 4/24 Applications to computational biology (guest lecture by Prof. Olga Troyanskaya) scribe notes (draft); slides
Th, 4/26 Applications to computer vision (guest lecture by Prof. Fei-Fei Li) scribe notes ; slides
Tu, 5/1 Computational neuroscience and fMRI data (guest lecture by Prof. Ken Norman) scribe notes ; slides;
"Beyond mind-reading: multi-voxel
pattern analysis of fMRI data"
Th, 5/3 Topic models: Hidden variable models for large document collections (and class summary) (DB) scribe notes ; slides

Books

Here is a list of optional books for further background reading. All of these are being placed on reserve at the Engineering Library.