CSE 446 - Machine Learning (original) (raw)
| Instructor: Luke Zettlemoyer (lsz at cs) Office hours: Fri 12:30-1:30, CSE 658 | TA: Antoine Bosselut (antoineb at cs) Office hours: Mon and Wed 10:30-11:30, CSE 220 TA: Naozumi Hiranuma (hiranumn at cs) Office hours: Tues 10:30-11:30, CSE 220 and Thur 1:30-2:30, CSE218 |
|---|
Schedule [_subject to change_]
| Week | Dates | Topics & Lecture Notes | Readings |
|---|---|---|---|
| 1 | Jan 5, 7, 9 | Introduction; Decision Trees | Murphy 1.1-1.4; 16.2 |
| 2 | Jan 12, 14, 16 | Point Estimation; Linear Regression | Murphy 6.4; 7.1-7.3; 7.5.1 |
| 3 | Jan 21, 23 | Linear Regression (cont.); Naive Bayes | Murphy 3.5; Murphy 8.1-8.3 |
| 4 | Jan 26, 28, 30 | Naive Bayes (cont.) Logistic Regression | Murphy 8.1-8.3 |
| 5 | Feb 2, 4, 6 | Logistic Regression(cont.); Perceptron | Murphy 8.5.0, 8.5.4 |
| 6 | Feb 9, 11, 13 | Kernels; Support Vector Machines (SVMs) | Murphy 14.5 |
| 7 | Feb 18, 20 | SVM (cont.) Boosting | Murphy 11.2, 11.4-11.4.2, 16.4 |
| 8 | Feb 23, 25, 27 | Clustering and EM | Murphy 12.1.0 |
| 9 | Mar 2, 4, 6 | Dimensionality Reduction; Instance-based Learning | Murphy 12.2 |
| 10 | Mar 9, 11, 13 | Learning Theory; Neural Networks (NNs) | Murphy 10.1 - 10.2.2 |
Textbooks
- Kevin Murphy, Machine Learning: a Probabilistic Perspective, MIT Press, 2013.
- Optional: Christopher Bishop, Pattern Recognition and Machine Learning, Springer, 2007.
- Optional: Tom Mitchell, Machine Learning, McGraw-Hill, 1997.
- Optional: R. Duda, P. Hart & D. Stork, Pattern Classification (2nd ed.), Wiley, 2001.
Contact
- The quickest way to contact us is to post to the GoPost discussion board. Posting your questions there will allow everyone to benefit from reading the answer. We also encourage you to try to answer questions, which will count towards class participation. We will monitor at least daily.
- Please feel free to email the course staff about topics not appriorate for the message board, although responses may not be as fast as on the board. Also, please come to office hours for instant feedback. Let us know if you need to meet outside of the scheduled hours, we will do our best to accomodate.
- Grades: Assignment grades are posted in the online CSE 446 Gradebook. Please let us know if you see any errors.
Homeworks
We will have 4 homework assignments, which will be listed below as they are assigned. The assignments will be given out roughly in weeks 2, 4, 6, and 8, and you will have two weeks to complete each one.
- Assignment 1: Decision Trees [pdf] [tex] [data] (Due Wed Jan 28th, 9:30am)
- Assignment 2: Classifiers: Naive Bayes, Perceptron, Logistic Regression [pdf] [data] (Due Fri Feb 13, 9:30am)
- Assignment 3: SVMs and Ensembles [pdf][data] (Due Fri Feb 27, 9:30am)
- Assignment 4: k-Means, EM, and PCA [pdf][data] (Due Fri Mar 13, 9:30am) Please submit your writeup and code to the online DropBox. Please also submit a printed copy of your writeup only (no need to print the code) at the beginning of class on the due date.
Exam
The final exam is Wednesday, March 18, 830-1020, in MOR 220. Here is a practice exam, with and without solutions. (ignore problem 13). The exam is open, you are welcome to bring the book, the lecture slides, and any handwritten notes you have.
Grading
The final grade will consist of homeworks (70%), a final exam (25%), and course participation (5%).
Course Administration and Policies
- Assignments will be done individually unless otherwise specified. You may discuss the subject matter with other students in the class, but all final answers must be your own work. You are expected to maintain the utmost level of academic integrity in the course.
- As we sometimes reuse problem set questions from previous years, please do not to copy, refer to, or look at any solution keys while preparing your answers. Doing so will be regared as cheating. We expect you to want to learn and not google for answers.
- Assignments may be handed in up to three days late, at a penalty of 15% of the maximum grade per day.
- Comments can be sent to the instructor or TA using this anonymous feedback form. We take all feedback very seriously and will do whatever we can to address any concerns.
 |
Department of Computer Science & Engineering University of Washington Box 352350 Seattle, WA 98195-2350 (206) 543-1695 voice, (206) 543-2969 FAX |