Stat 406 Spring 2007

• Lectures MWF 1-2, Math building 105
• Lab: Wed 4-5pm, Stat 310
• Office hours: Wed 2pm, Stat 308d
• Instructor: Kevin Murphy
• TA: Virginia Chen
• Syllabus
• Grading: Midterm: 30%, Final: 45%, Weekly ssignments: 25%
• Textbook. Required: Pattern Recognition and Machine Learning, Christopher M. Bishop, Springer 2006.
  Recommended: The Elements of Statistical Learning, Trevor Hastie, Robert Tibshirani, Jerome Friedman, Springer 2001.
  Recommended: All of statistics, Larry Wasserman, Springer 2004
  Recommended: Pattern Classification (2nd ed.), Duda, Hart, Stork, Wiley 2001.
  We will also use handouts.
• pre-requisites: cpsc 340 (machine learning) or stat 306 (finding relationships in data). You should be familiar with linear algebra, multivariate calculus, probability theory and basic programming (we will use Matlab).

News

• pcaFast.m added
• Final date announced : April 27th
• google groups created, email ubcstat406spring07@googlegroups.com

Timetable

L#	Date	Slides	Reading	Homework
L1	Mon Jan 8	lec1.pdf	B 1.1-1.2	hw1.pdf
L2	Wed Jan 10	Intro contd	Matlab tutorial.pdf More Matlab tutorials	lab1.pdf, lab1Data.zip, Clarifications,
L3	Fri Jan 12	Bayesian classifiers for Gaussian data	Generative classifiers, B1.5, B2.1, B2.2, B4.2	.
L4	Mon Jan 15	cont'd	M 12.1	hw2.pdf Download Code.zip, CodeEx.zip, Data.zip. You will need these files for this and future homeworks.
L5	Wed Jan 17	MLEs for 1D Gauss and Multinomial	M 12.2.1, M 4.2.3.1, M.2.3.2, M 4.2.3.4	.
L6	Fri Jan 19	MLEs for MVN	M 12.2.2-12.2.3, M 4.2.3.5	.
L7	Mon Jan 22	Decision boundaries	M 12.2.5	.
L8	Wed Jan 24	Naive Bayes	M 12.3	hw3.pdf
L9	Fri Jan 26	Classifiers with missing data/ Marginals and conditionals of a MVN	M 3.4.2.1, 3.4.2.7, 3.4.2.10	.
L10	Mon Jan 29	Bayesian estimation of a 1D Gaussian mean	M 5.3.1	.
L11	Wed Jan 31	Shrinkage estimate of 1D Gaussian mean	M 5.7.1	.
L12	Fri Feb 2	cont'd	.	hw4.pdf due Monday 12th. You will also need gaussClassifMissingData.mat and Monday's handout (see L13)
L13	Mon Feb 5	Beta-binomial model	M 5.2	.
L14	Wed Feb 7	EB for BB model	Empirical Bayes for Beta-Binomial model	.
L15	Fri Feb 9	Netflix (guest lecture from Hoyt Koepke)	.	.
L16	Mon Feb 12	Midterm Review		.
L17	Wed Feb 14	Review cont'd	.	.
L18	Fri Feb 16	.	.	.
L19	Mon Feb 19	Spring break	.	.
L20	Wed Feb 21	Spring break	.	.
L21	Fri Feb 23	Spring break	.	.
L22	Mon Feb 26	Midterm	.	.
L23	Wed Feb 28	Least squares	Linear regression	.
L24	Fri Mar 2	Ridge regression	.	hw5.pdf, sinusoidData.mat, standardizeCols.m, degexpand.m,
L25	Mon Mar 5	Ridge regression	.	.
L26	Wed Mar 7	Ridge regression	.	.
L27	Fri Mar 9	Generalized ridge regression	.	.
L28	Mon Mar 12	Bayesian linear regression	.	hw6.pdf, pcaFaceDemo.m, sqdist.m, pcaHighDim.m, faceOlivetti_trainTest.mat, faceOlivetti_trainTestV6.mat (for people who do not have Matlab 7)
L29	Wed Mar 14	PCA	pcaHandout.pdf	.
L30	Fri Mar 16	Probabilistic PCA	.	.
L31	Mon Mar 19	PPCA	.	hw7.pdf, code.zip pcaFast.m,
L32	Wed Mar 21	No class	.	.
L33	Fri Mar 23	No class	.	.
L34	Mon Mar 26	Latent Semantic Indexing	LSI	.
L35	Wed Mar 28	Gaussian mixture models	Mixture models	.
L36	Fri Mar 30	EM	.	hw8.pdf, lsiDocuments.pdf, lsiMatrix.txt, lsiWords.txt
L37	Mon Apr 2	K means	.	.
L38	Wed Apr 4	EM for Bernoulli mixtures	.	.
L39	Fri Apr 6	Good Friday	.	.
L40	Mon Apr 9	Easter Monday	.	.
L41	Wed Apr 11	Review	.	.

Final: April 27th, 12 noon, Math 105.