CS 534 - Machine Learning [Spring 2017]

Schedule

#	Date	Topic	Materials	References	Assignments
Introduction and Review
1	1/10	Overview & Course Logistics	Lecture 1 iPythonIntro.ipynb	Chapter 1 (Hastie et al.) Chapter 1 (Murphy)
2	1/12	Random Variables & Probability Review	Lecture 2 Bayesian Review	Probability theory review from Stanford's machine learning class Sam Roweis's probability review notes Ingmar Land's probability cheat sheet
3	1/17	Linear Algebra & Optimization Review	Lecture 3 NumpyBasics.ipynb	Linear algebra notes from Stanford's ML class Sam Roweis's linear algebra review notes Convex optimization notes Part I and II from Stanford's machine learning class
Supervised Learning I
4	1/19	Statistical Decision Theory & Linear Regression	Lectures 4 - 5 PandasBasics.ipynb Teams.csv (data file) LinearRegression.ipynb	Chapter 2.1 - 2.4 (Hastie et al.) Chapter 3.1 - 3.4 (Hastie et al.) Chapter 17.1 - 17.2 (Barber) Prof. Carlos Carvalho MLR Slides
5	1/24	Statistical Decision Theory & Linear Regression
6	1/26	Linear Classification	Lectures 6-7 LDA.ipynb LogisticRegression.ipynb	Chapter 4.1 - 4.4 (Hastie et al.) Chapter 3.2 - 3.5 (Mitchell)	Homework 1 Spam Training Data Spam Testing Data
7	1/31	Linear Classification	Lectures 6-7 LDA.ipynb LogisticRegression.ipynb
Validation, Model Selection, and Theory
8	2/2	Learning Theory	Lecture 8	Chapters 7, 8 (Hastie et al.) Learning theory notes from Stanford's ML class Chapter 12 (Daume)
9	2/7	Validation	Lecture 9 CrossValidation.ipynb
10	2/9	Bootstrap & Model Selection	Lectures 10-11		Homework 2
11	2/14	Bootstrap & Model Selection	Lectures 10-11
Supervised Learning II
12	2/16	Boosting, Trees & Additive Models	Lectures 12-13 DecisionTree.ipynb	Chapters 9, 10 (Hastie et al.) Chapter 3 (Mitchell) Cynthia Rudin's notes on decision trees
13	2/21	Boosting, Trees & Additive Models	Lectures 12-13 DecisionTree.ipynb
14	2/23	Ensembles & Random Forests	Lecture 14	Chapters 15, 16 (Hastie et al.) Breiman's paper on random forests
15	2/28	Support Vector Machines	Lectures 15-16 SVM.ipynb	Chapter 12 (Hastie et al.) Chapter 15 (Shalev-Shwartz & Ben-David) SVM notes from Stanford's ML class SVM notes from NYU's ML class Chapter 11 (Daume)	Homework 3 allhyper.data
16	3/2	Support Vector Machines	Lectures 15-16 SVM.ipynb
Supervised Learning II
17	3/14	Neural Networks	Lectures 17-18 Neural Network Playground Handwritten Digit Visualization plot_mnist_filters.ipynb by Scikit-learn	Chapter 11 (Hastie et al.) Notes from Convolutional Neural Networks course at Stanford Chapter 1, 2 (Nielsen)
18	3/16	Neural Networks
19	3/21	K Nearest Neighbors	Lecture 19	Chapter 13 (Hastie et al.)	Homework 4
Midterm
20	3/23	Midterm
Unsupervised Learning
21	3/28	Dimensionality Reduction	Lecture 21 DimensionalityReduction.ipynb	Chapter 14 (Hastie et al.) Chapter 23 - 23.1 (Shalev-Shwartz & Ben-David) PCA notes from Stanford's ML class
22	3/30	Clustering & Mixture Models	Lecture 22
Other Topics
23	4/4	Topic Models	Lecture 23	Blei's review article	Homework 5 Colleges.txt
24	4/6	Hidden Markov Models	Lecture 24	Murphy's intro to graphical models HMM tutorial by Blunsom
25	4/11	Deep Learning	Lecture 25
26	4/13	Recommendation Systems	Lecture 26	Chapter 9 (Leskovec et al.) Paper by Koren et al.
Project Presentations
27	4/18	Presentations
28	4/20	Presentations

Course Logistics

Piazza

All announcements, assignment clarifications, and slide corrections will be posted on Piazza. Make sure to check the site on a regular basis.

Office Hours

Joyce Ho: M 1:30 PM - 3:30 PM, W 9:30 AM - 12:00 PM @ MSC W414
Rongmei Lin: F 12:00 PM - 2:00 PM @ MSC N410

Note: Office hours may change from time to time, in which case an announcement will be made on Piazza.

Textbook

Required: The Elements of Statistical Learning: Data Mining, Inference, and Prediction), by Trevor Hastie, Robert Tibshirani & Jerome Friedman
Supplemental: Machine Learning: a Probabilistic Perspective, by Kevin Murphy
Supplemental: Pattern Recognition and Machine Learning, by Christopher Bishop

Course Grading

Component	Weight
Homeworks	40%
Midterm	15%
Project	40%
Participation	5%

Assignment & Exam Policies

Assignments:

Assignments are due electronically at 11:59 PM on Canvas.
Each student receives six late days that can be used on assignments throughout the semester. These late days extend the deadline for 24 hours, and can be distributed amongst the 4-5 assignments, but no more than 3 late days may be used on any assignment.
After the six late days are used, the assignment will be penalized 10% off per day up to 3 late days.
Late days apply to the entire assignment, so handing in one problem late counts as a late day towards the whole assignment.
After 3 late days on any given assignment you will receive no credit for the asignment.

Exams:

All exams must be taken promptly at the required time.
Requests for rescheduling the midterm exam will only be considered if the request is made at least a week prior to the exam date.

The above policies will be waived only in an "emergency" situation with appropriate documentation from OUE.

Honor Code

All class work is governed by the College Honor Code and Departmental Policy. It is acceptable and encouraged to discuss assignments with other students but any code and writeups must be written by yourself.