Machine Learning CSCI 5622
Fall 2006.
Course Syllabus
Homeworks:
·
Homework 1: Equations from Code
(Assigned: September 6, 2006. Due:
September 20,
2006). Matlab Code (zipped).
This assignment is worth 3% of your final mark.
·
Homework 2: Matlab Code From
Equations (Assigned: September 13, 2006. Due:
September 27,
2006). Matlab code: Demo_Gaussian_Kernel.m,
Demo_Poly_Kernel.m, and Demo_Sigmoid_Kernel.m. This assignment is
worth 3% of your final mark.
·
Homework 3: Kernel Ridge
Regression (Assigned: September 20, 2006. Due:
October 11, 2006).
This assignment is worth 10% of your final mark.
- Data: The data for this
homework is in X_lrn_HW3.txt and Y_lrn_HW3.txt. Example Matlab code for reading this
data is in Read_HW3_Data.m.
·
Homework 4: Implementing a
Learning Algorithm from Equations and Learning to use SVMs. (Assigned:
October 11, 2006. Due:
November 15, 2006). This assignment is worth 29% of your final
mark. Data X_4_trn.txt, Y_4_trn.txt,
X_4_val.txt, Y_4_val.txt, and
X_4_tst.txt. Hints file.
.
Quiz:
·
Take home quiz 1: Regression
(Assigned: September 13, 2006. Due:
September 20,
2006 at start of class). This quiz is worth 1% of
your final mark.
Weekly
Class Schedule:
- August 30: Introduction.
Class Survey Survey.txt (email to me by end of
week).
- September 6: Introduction
to Regression. Ridge Regression Summary.
Homework 1 Assigned – Due on September 20
- September 13: Introduction
to Classification. Assumptions and
Definitions. Homework 2 Assigned – Due
on September 27
- September 20: Perceptron
Algorithm. Model Selection and
Estimating Model Error. Homework 3
Assigned – Due on October 11
- September 27: Support
Vector Machine Classification.
- October 4: Support
Vector Machine Regression.
- October 11: Neural
Networks. Nearest
Neighbor Classification and Regression. Homework
4 Assigned – Due on November 1
- October 18: Classification and
Regression Trees.
- October 25: Ensemble learning (Boosting.
Bagging, Random Forests).
- November 1: Linear and Quadratic Discriminant
Classification Models. Density Estimation with Mixture of Gaussians. Data for class project posted.
- November 8: Supervised Learning
Summary.
- November 15: Midterm
- November 22: No Class
- November 29: Clustering, Spectral Clustering,
Semi-Supervised Learning
- December 6: Reinforcement Learning.
- December 13: Summary.