| Philippos Mordohai Assistant Professor Department of Computer Science Stevens Institute of Technology Office: Lieb 215
|
CS 559: Machine Learning: Fundamentals and Applications Spring 2010 |
Homepage |
Location Burchard 124. Note that Burchard 124 has a separate entrance to the left of the main entrance to Burchard. Time Thursdays 6:15-8:45 PM. Office Hours Tuesday 4:30-6 and by e-mail. Pre-requisites Basic knowledge of probability and statistics. Past experience has shown that students without this background struggle in CS 559. Basic programming in Matlab, C/C++ or Java. This is crucial for the final project which requires the implementation of machine learning techniques. Syllabus Textbook The required textbook is the following. I refer to it as DHS in the class outline. I will also use notes outside the textbook, mostly in the second half of the semester. Pattern Classification (2nd Edition) (Hardcover) by Richard O. Duda, Peter E. Hart and David G. Stork Publisher: Wiley-Interscience; 2 edition (October 2000) ISBN-10: 0471056693 ISBN-13: 978-0471056690 Errata, slides (some of which I will use during the class) and other information can be found at the book website Slides for each lecture will be posted here before the lecture. Evaluation Project (25%) Each student will select a project related to machine learning, which has to be approved by me regarding relevance and feasibility. I will provide pointers and suggestions for potential projects. Students actively involved in machine learning research can select a project related to their research, but new work has to be done during the semester. Large projects can be performed by groups of two students. Each student will briefly present his or her project in 3-5 minutes during Week 9. Final project reports and presentations are due in Week 14. 4 homework sets (20%) Homework sets will be tentatively assigned in Weeks 3, 5, 10 and 12 and will be due a week later. The penalty for late submission is 20% of the grade per day. Pop-up quizzes and participation (10%) A simple quiz will be given at the beginning of each class. Midterm (20%) The midterm is scheduled for Week 8 (March 11). It should only take the first hour of the class. Final (25%) The final will take place during the final exam period and will be cumulative. Outline Notes pt. 1 (pdf) Week 2: Bayesian decision theory (DHS Ch. 2) Notes pt. 2 (pdf) Week 3: Maximum likelihood estimation and Bayesian parameter estimation (DHS Ch.3) Notes pt. 3 (pdf) Homework 1 (pdf) is due Feb. 11. Week 4: Expectation Maximization (DHS Ch.3 + notes) Notes pt. 4 (pdf) Week 5: Expectation Maximization (part II) and Hidden Markov Models (DHS Ch.3 + notes) Notes pt. 5 (pdf) Homework 2 (pdf) is due Feb. 25. Week 6: Hidden Markov Models and Principal Component Analysis(DHS Ch. 3 and 4, notes) Notes pt. 6 (pdf) (includes links to datasets for projects). Notes will be covered in Week 7. PCA functions in Matlab (pdf). Week 7: Eigenfaces and Fisher Linear Discriminant (DHS Ch. 3 and notes). Notes pt. 7 (pdf) (recap up to now). Lecture based on Week 6 notes. Week 8: Midterm and Nonparametric techniques (DHS Ch. 5 + notes) Notes pt. 8 (pdf) Week 9: Project proposals, Linear Discriminant Functions and Perceptron (DHS Ch. 5) Notes pt. 9 (pdf) Homework 3 (pdf) is due April 8. Week 10: MSE Procedures for Linear Discriminat Functions and Support Vector Machines (DHS Ch. 5) Notes pt. 10 (pdf) Week 11: Boosting, bagging and random forests (notes) Notes pt. 11 (pdf) Homework 4 (pdf) is due April 22. Week 12: Unsupervised Learning and Clustering (Notes and DHS Ch.10). Notes pt. 12 (pdf) Week 13: Project presentations Notes pt. 13 (pdf) (recap of second part) |