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

EEL6825: Tentative Course Outline

TENTATIVE EEL 6825 COURSE OUTLINE

08/22 L01 Introduction
08/24 L02 Bayes Decision Rule, 1D example, Bayes error is optimal

08/27 L03 Univariate Normal, multivariate, covariance matrix,
           eigenvectors/values
08/29 L04 Mahalanobis distance, plotting equiprob distributions, 
           linear transforms.
08/31 L05 General quadratic decision boundaries
           sigma_i same => linear boundaries

09/03 LABOR DAY
09/05 L06 Review HW#1 questions,
          How to compute and estimate error, Bhatacharrya bound
09/07 L07 HW#1 due, Generating random Gaussian data,
          review linear transforms, Orthonormal transform, 
          distance is preserved

09/10 L08 whitening, distance is not preserved
          simultaneous diagonalization
09/12 L09 Maximum Liklihood estimates of mean and variance
          bias, consistency
09/14 L10 Finish max liklihood, MAP

09/17 L11 Bayesian Estimation
09/19 L12 HW#2 questions, Introductions to Linear Classifiers, 
          geometric interp.
09/21 L13 HW#2 due, Fisher Criterion
           extend Fisher to multi-dim
           nonlinear features
           quadratic classifier

09/24 L14 Linear classifiers
09/26 L15 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?
09/28 L16 Review for Exam 1

10/01     NO CLASS--Exam 1, in NEB101, period E1-E3 (7:20pm-10:10pm)
10/03 L17 Parzen Windows       
10/05 L18 1-NN

10/08 L19 1-NN, K-NN
10/10 L20 error of 1-NN
10/12 Veteran's Day

10/15 L21 Go over Exam I
10/17 L22 HW#3 questions, Projects handout
10/19 L23 HW#3 due, begin to motivate neural networks, perceptron, 
          linear system solution

10/22 L24 single-layer perceptron
10/24 L25 multi-layer perceptrons
10/26 L26 backpropagation learning

10/29 L27 KL dimensionality reduction
10/31 L28 KL dimensionality reduction 
11/02 HOMECOMING, NO CLASS

11/05 L29 HW#4 questions, KL summary and examples
11/07 L30 HW#4 due, Alternative Dimensionality Reduction Methods
11/09 L31 HW#4 due, Review for Exam 2

11/12      NO CLASS--Exam 2, in NEB202, period E1-E3 (7:20pm-10:10pm)
11/14 L32 Unsupervised learning and clustering
11/16 L33 Speech applications

11/19 L34 Go over Exam 2
11/21 THANKSGIVING VACATION, NO CLASS
11/23 THANKSGIVING VACATION, NO CLASS

11/26 L35 Project questions 
11/28 L36 FINAL PROJECTS
11/30 L37 FINAL PROJECTS

12/03 L38 FINAL PROJECTS
12/05 L39 FINAL PROJECTS



Dr John Harris 2001-11-26