next up previous
Next: EEL6825: Projects Up: Administration Previous: EEL6825: Mailing List

EEL6825: Course Outline

TENTATIVE EEL 6825 COURSE OUTLINE

AUG 23 L01 Introduction
AUG 25 L02 Bayes Decision Rule, 1D example, Bayes error is optimal
AUG 27 L03 Univariate Normal, multivariate, covariance matrix,
           eigenvectors/values

AUG 30 L04 Mahalanobis distance, plotting equiprob distributions, 
           linear transforms.
SEP 01 L05 General quadratic decision boundaries
           sigma_i same => linear boundaries
SEP 03 L06 How to compute and estimate error, Bhatacharrya bound

SEP 06     NO CLASS -- LABOR DAY
SEP 08 L07 Review HW#1 questions, generating random Gaussian data,
           review linear transforms, Orthonormal transform, 
           distance is preserved
SEP 10 L08 HW #1 Due, 
           whitening, distance is not preserved
           simultaneous diagonalization

SEP 13 L09 Maximum Liklihood estimates of mean and variance
           bias, consistency
SEP 15     NO CLASS -- HURRICANE FLOYD!
SEP 17 L10 Finish max liklihood, MAP

SEP 20 L11 Bayesian Estimation
SEP 22 L12 Introductions to Linear Classifiers, geometric interp.
SEP 24 L13 HW#2 due, Fisher Criterion
           extend Fisher to multi-dim
           nonlinear features
           quadratic classifier

SEP 27 L14 Marc Boillot guest lecture
SEP 29 L15 Linear classifiers
OCT 01 L16 resubstitution, hold-out, leave-one-out
           test set vs. training set
           bad use of test and training
           How much data do you need?
           how much data for classifier, for estimating error?

OCT 04 L17 Review for exam 1
OCT 06     Exam I
OCT 08 L18 Parzen Windows

OCT 11 L19 Parzen Windows, 1-NN
OCT 13 L20 1-NN, K-NN
OCT 15 L21 Go over Exam I

OCT 18 L22 Neural Networks 1
OCT 20 L23 Neural Networks 2 (Principe)
OCT 22 L24 HW#3 due (KL, Parzen,k-NN) due, Neural Networks 3 (Principe)

OCT 25 L25 Neural Networks 4 (Principe)
OCT 27 L26 Neural Networks 5 (Principe)
OCT 29 L27 Neural Networks 6 (Principe)

NOV 01 L28 Finish Neural Networks, discuss final projects
NOV 03 L29 KL dimensionality reduction
NOV 05 NO CLASS Homecoming

NOV 08 L30 KL dimensionality reduction summary and examples
NOV 10 L31 HW questions
NOV 12 L32 HW #4 DUE (neural networks, K-L), individual meetings

NOV 15 L33 Review for Exam II      
NOV 17     EXAM II
NOV 19 L34 speech processing

NOV 22 L35 Hand back exams
NOV 24     NO CLASS THANKSGIVING
NOV 26     NO CLASS THANKSGIVING

NOV 29 L36 FINAL PROJECTS
DEC 01 L37 FINAL PROJECTS
DEC 03 L38 FINAL PROJECTS

DEC 06 L39 FINAL PROJECTS
DEC 08 L40 FINAL PROJECTS



Dr John Harris
1999-12-10