CS 445/545
Spring Quarter 2017
Time : Tuesdays and Thursdays, 2:003:50pm
Location: ASRC 215
Instructor:
Melanie Mitchell,
FAB 11513, (503) 7252412, mmATpdx.edu
Office hours: M,W 11amnoon, or by appointment.
Teaching Assistant:
Mike Lane, lane7ATpdx.edu
Office hours: T,Th Noon1pm, FAB 115 in the "Data Lab" (across the hall from 11513).
Course Mailing List: mlspring2017@cs.pdx.edu
Prerequisites: Undergraduatelevel courses in calculus, linear algebra, and probability and statistics. Facility in at least one highlevel programming language.
Main topics:: Perceptrons, neural networks, linear regression, logistic regression, evaluating classifiers, support vector machines, decision trees, ensemble learning, Bayesian learning, unsupervised learning, reinforcement learning, and deep learning.
Textbook: No textbook. Readings will be assigned from materials available online.
Relation to CS 441/541 (Artificial Intelligence): A couple of the same topics will be covered (e.g., neural networks), but these will be covered in more depth and at a more theoretical level in this course than in CS 441/541. Otherwise, the topics in this course differ from the topics covered in CS 441/541.
Homework: The class will have several homework assignments, involving writing code for and/or experimenting with various machine learning methods, along with written exercises.
Late homework policy: Students must request and be granted an extension on any homework assigment before the assignment is due. Otherwise, 5% of the assignment grade will be subtracted for each day the homework is late, up to a maximum of 25%.
Exams: The class will have an inclass final exam. There will also be several short inclass quizzes to test basic understanding of the material presented in class and in the readings. You are allowed to bring in one doublesided page of notes for each quiz, and four doublesided pages of notes for the final exam.
Grading: Homework 50%, Quizzes 20%, Final exam 30%.
Academic integrity: Students will be responsible for following the PSU Student Conduct Code.
Students with disabilities: If you are a student with a disability in need of academic accommodations, you should register with the Disability Resource Center and notify the instructor immediately to arrange for support services.
Syllabus (subject to change):
Date 
Topics 
Homework and Reading 
Tuesday April 4 
Introduction to machine learning (pptx or pdf) Class "To Do" list Pretest solutions 
Reading: 
Thursday April 6 
Perceptrons, continued 
Homework 1 (Perceptrons), due Tuesday April 18 Assignment, mnist_train.csv, mnist_test.csv 
Tuesday April 11 


Thursday April 13 
Multilayer neural networks, continued Partial solutions to inclass exercises Review for Quiz 1 

Tuesday April 18 
Quiz 1 Quiz 1 solutions Evaluating classifiers (pptx or pdf) Inclass exercises solutions (partial) 
Reading:: T. Fawcett, An introduction to ROC analysis, Sections 14, 7 Homework 2 (Neural Networks) , due Thursday April 27: Assignment Update: Homework 2 deadline is now extended to Monday, May 1, 5pm 
Thursday April 20 
Reading: A. BenHur and J. Weston, A User's Guide to Support Vector Machines  
Tuesday April 25 
Support vector machines, continued Feature selection Review for Quiz 2 

Thursday April 27 
Quiz 2 Quiz 2 solutions 
Reading: Tom Mitchell, Generative and discriminative classifiers: Naive Bayes and logistic regression, Sections 12. Homework 3 (SVMs and feature selection) , due Tuesday May 9: Assignment 
Tuesday May 2 
Bayesian learning, continued  
Thursday May 4 
Bayesian learning, continued 

Tuesday May 9 
Review for Quiz 3 Logistic regression (pptx or pdf) 
Homework 4 (Bayesian learning and logistic regression), due Thursday May 18: Assignment Reading: X. Z. Fern, Notes on Logistic Regression 
Thursday May 11 
Quiz 3 Quiz 3 solutions Ensemble learning, continued 
Reading: R. Schapire, A brief introduction to boosting 
Tuesday May 16 
Reading: Chapter 8 from Introduction to Data Mining by Tan, Steinbach, and Kumar, pp. 487515, 532540, 546550. 

Thursday May 18 
Unsupervised learning, continued Review for Quiz 4 
Homework 5 (Kmeans clustering), due Tuesday May 30: Assignment Optdigits dataset for Homework 5 
Tuesday May 23 
Quiz 4 Quiz 4 solutions 
Reading: pp. 278280 of Reinforcement Learning for True Adaptive Traffic Signal Control (a brief and clear introduction to Qlearning) 
Thursday May 25 
Reinforcement learning, continued 

Tuesday May 30 
Review for Quiz 5 Deep learning (pptx or pdf) 
Homework 6 (Reinforcement learning), due Thursday June 8 : Assignment 
Thursday June 1 
Quiz 5 Quiz 5 solutions Deep learning, continued 

Tuesday June 6 
Review for final exam 

Thursday June 8 
Guest lectures 

Monday June 12 
Final exam: 10:15am  12:05pm 
