HW6: Due March 23, 2009

Problems 4.4, 4.5, 4.9, 4.10, 4.12, 4.13, 4.15

Build logistic regression classifiers for the Ellipse, Cross, and Iris data (which will be made available through elearning) using the IRLS training algorithm

HW5:

Repeat the experiments (qualitatively) in Figures 3.7, 3.8, and 3.9 preferably using MATLAB. For the case of 3.8, verify empirically that the “posterior distribution becomes narrower” as additional data points are observed. Please submit your homeworks to cap6610sp09@gmail.com.

For on campus students, please submit by Feb 11(W) 11:59pm, 2009. For edge students, please submit by Feb 18(W), 11:59pm, 2009.

HW4: 2.38, 2.40, 2.44, 3.2, 3.3 (Bishop)

Please submit your homeworks to cap6610sp09@gmail.com

Please include your Name, UFID, and Email in your homework, also please indicate whether you are On campus or Edge students.

Thanks.