Introduction To Machine Learning
Spring 2016
Overview Machine learning is an exciting and fastmoving field of
computer science with many recent consumer
applications (e.g., Microsoft Kinect, Google Translate,
Iphone's Siri, digital camera face detection, Netflix
recommendations, Google news) and applications within the
sciences and medicine (e.g., predicting proteinprotein
interactions, species modeling, detecting tumors,
personalized medicine). In this undergraduatelevel class,
students will learn about the theoretical foundations of
machine learning and how to apply machine learning to
solve new problems. 

General information Lectures: Tuesday and Thursday,
2pm3:15pm
Office hours (David): Tuesdays
4:305:30pm. Location:
715 Broadway, 12th floor, Room 1204 Grading: problem sets (50%) + midterm
exam (25%) + project (20%) + participation (5%). Problem Set policy Prerequisites: Students must
either have taken Basic Algorithms (CSCIUA.0310) or be
taking it concurrently. Linear algebra (MATHUA 140) is
strongly recommended as a prerequisite, and knowledge of
multivariable calculus will be helpful. Students should
also have good programming skills.
Piazza:
We will use Piazza to answer
questions and post announcements about the course.
Please sign up here. Students' use of Piazza,
particularly for adequately answering other students'
questions, will contribute toward their participation
grade. 
Schedule
Lecture  Date  Topic  Required reading  Assignments 
1  Jan 26 (Tues) 
Overview [Slides]

Chapter
1 of Murphy's book 

2  Jan 28 (Th) 
Introduction to learning [Slides] Loss functions, Perceptron algorithm, proof of perceptron mistake bound 
Barber 17.1 on
leastsquares regression, A.1.14 (review of vector
algebra) Notes on perceptron mistake bound (just section 1) 
ps1 (data), due Feb 5 at 6pm 
3 
Feb 2 (Tues) 
Linear
classifiers [Slides] Introduction to Support vector machines 


4 
Feb 5 (Thurs) 
Support
vector machines [Slides] [iPython notebook, html] Introduction to convex optimization, gradient descent 
Notes on support
vector machines (sections 14) Additional notes on SVMs (sec. 1 & 2) Notes on optimization 
ps2, due Feb 15 at 10pm. [Solutions] 
5 
Feb 9 (Tues) 
Stochastic
gradient descent [Slides] Pegasos algorithm (stochastic subgradient descent for SVMs) 

6 
Feb 11 (Thurs) 
Kernel
methods [Slides] Kernel methods for SVMs, multiclass classification 
Notes on kernels
(section 7) Lecture notes Optional: ShalevShwartz & BenDavid Chapter 16 on kernel methods Optional: ShalevShwartz & BenDavid Sections 17.1 & 17.2 on multiclass 

7 
Feb 16 (Tues) 
Kernel methods (continued) 
Python demo shown
in class 
ps3 (data), due
Feb 24 at 10pm 
8 
Feb 18 (Thurs) 
L1regularization + Diabetes case study
[Slides] Intro to learning theory 
Optional: Diabetes
paper 

9 
Feb 23 (Tues) 
Learning theory [Slides] Finite hypothesis classes 
Notes
on learning theory (sections 13) 

10 
Feb 25 (Thurs) 
Learning theory [Slides] VCdimension 
Notes
on learning theory (section 4) Optional: Notes on gaptolerant classifiers (section 7.1, pg. 2931) 
ps4 (data), due Mar 6 at 10pm 
11 
Mar 1 (Tues) 
Decision trees [Slides] Ensemble methods 
Mitchell
Ch. 3 Optional: Hastie et al., Section 8.7 (bagging) Optional: Rudin's lecture notes (on decision trees) Optional: Hastie et al. Chapter 15 (on random forests) 

12 
Mar 3 (Thurs) 
Kmeans clustering [Slides] 
ShalevShwartz
& BenDavid Chapter 22 intro and Section
22.2 Optional: Hastie et al., Sections 14.3.6, 14.3.8, 14.3.9 
ps5 (data) due Mar 21 at
10pm 
13 
Mar 8 (Tues) 
Hierarchical & spectral clustering [Slides] 
ShalevShwartz
& BenDavid Sections 22.1, 22.3 Hastie et al., Sections 14.3.12, 14.5.3 Optional: Tutorial on spectral clustering 

14 
Mar 10 (Thurs) No office hours during spring break 
Introduction to Bayesian inference
[Slides] Bayes rule, decision theory 
The
Go Files: AI computer wins first match against master Go
player Optional: Silver et al. Nature article (NYU access) 

15 
Mar 22 (Tues) 
Midterm review 

Mar 24 (Thurs) 
Midterm exam (in class)  Project proposal
due Mar 28 at 10pm 

16 
Mar 29 (Tues) No office hours Mar 29 
Naive Bayes [Slides] Maximum likelihood estimation 
Notes
on naive Bayes (Sections 1 & 2) ShalevShwartz & BenDavid Chapter 24 (except 24.4) 

17 
Mar 31 (Thurs) 
Logistic regression [Slides] 
Notes on logistic regression (Sections 35)  
18 
Apr 5 (Tues) 
Graphical models [Slides] Modeling temporal data (e.g., hidden Markov models) 
Tutorial
on HMMs Introduction to Bayesian networks Optional: An introduction to graphical models 

19 
Apr 7 (Thurs) 
Unsupervised learning I [Slides] Expectation maximization 
Notes
on mixture models Notes on Expectation Maximization ShalevShwartz & BenDavid Section 24.4 

20 
Apr 12 (Tues) No office hours Apr 12 
History of artificial intelligence
[Slides] Guest lecture by Prof. Zaid Harchaoui 
A
(Very) Brief History of Artificial Intelligence Optional: Chapter 1 of Russell & Norvig (available at NYU libraries) 

21 
Apr 14 (Thurs) 
Unsupervised learning II [Slides] Topic models (e.g., latent Dirichlet allocation) 
Review article on topic
modeling Explore topic models of: politics over time, stateoftheunion addresses, literary studies (see also this blog), Wikipedia 
ps6 due
Apr 25 at 10pm 
22 
Apr 19 (Tues) 
Topic
modeling (continued) [Slides] 

23 
Apr 21 (Thurs) 
Dimensionality
reduction [Slides] Principal components analysis 
ShalevShwartz
& BenDavid Section 2323.1 Optional: Barber, Chapter 15 Optional: Notes on PCA Optional: More notes on PCA 

24 
Apr 26 (Tues) 
Introduction to neural networks [Slides] Backpropagation, convolution 
Notes on
backpropagation (extra) Optional: Neural network playground Optional: Nature article on deep learning 
ps7 (data for q2,
data
for q3) due May 9th at 10pm 
25 
Apr 28 (Thurs) 
Lab on deep learning TensorFlow tutorial by Yijun Xiao 

26 
May 3 (Tues) 
Project presentations (group 1) 

27 
May 5 (Thurs) 
Project presentations (group 2) 
Final
project writeup due May 17th at 12pm, via NYU classes 
Reference materials

Problem
Set policy I expect you to try solving each problem set on your own. However, when being stuck on a problem, I encourage you to collaborate with other students in the class, subject to the following rules:
