Tue, 2:00-3:40pm, Room: 19 West 4th St, Rm 101 (capacity = 40) Instruction Mode: Blended
3001.018: Wednesday from 9am-9:50am Instruction Mode: Online
3001.002:Wednesday from 3:30pm-4:20pm Room: 60FA_150 (capacity = 17) Instruction Mode: Blended
Cristina Savin, csavin@nyu.edu
Office hours: TBD (Online)
Ashwin Siripurapu (in-person for section 002 - blended), ars991@nyu.edu
Jiyuan Lu (remote for section 002 - blended), jl11046@nyu.edu
Colin Bredenberg (remote for section 018), cjb617@nyu.edu
Yiqiu (Artie) Shen, (grader) ys1001@nyu.edu
Office hours: TBD (Online)
This graduate level course presents fundamental tools for characterizing data with statistical dependencies over time, and using this knowledge for predicting future outcomes. These methods have broad applications from econometrics to neuroscience.The course emphasizes generative models for time series, and inference and learning in such models. We will cover range of approaches including Kalman Filter, HMMs, AR(I)MA, Gaussian Processes, and their application to several kinds of data.
Note: information presented is tentative, syllabus may be subject to change as course progresses.
problem sets (25%) + midterm exam (20%) + final project (25%) + lab(20%)+participation(10%)
Participation: piazza, engagement during lectures, labs, and office hours
We will use Piazza for announcements, and discussions about the course. Interactions on Piazza, particularly good answers to other students' questions, will count toward the participation grade.
Work in groups of 2-3 students.* Topics are flexible, including applying know algorithms to an interesting dataset, reviewing and implementing a state of the art solution, to improving an existing algorithm. Project proposals due in week 4.
*Check with CS if you are considering working individually or in a larger group.
Lecture videos will be posted to NYU Classes. Class attendance is still required.
| Date | Lecture | Assignments |
|---|---|---|
| Sept. 2 | No lab | |
| Sept. 8 | Lecture 1: Logistics. Introduction. Basic statistics for characterizing time series. | |
| Sept. 9 | No lab. Recap basic Bayes, graphical models as prerecorded video (classes meet on Mo schedule)] | |
| Sept. 15 | Lecture 2: AR basic inference and learning | |
| Sept. 16 | Lab 1: AR | |
| Sept. 22 | Lecture 3: ARIMA models | |
| Sept. 23 | Lab 2: ARIMA | |
| Sept. 29 | Lecture 4: LDS, Kalman filtering | |
| Sept. 30 | Lab 3: Inference in LDS | |
| Oct. 6 | Lecture 5: Particle filtering | |
| Oct. 7 | Lab 4: LSD parameter learning | |
| Oct. 13 | Lecture 6: Hidden Markov Models | Project proposal due |
| Oct. 14 | Lab 5: Particle filtering | |
| Oct.20 | Lecture 7: a unified view of linear models | |
| Oct.21 | Lab 6: HMMs | |
| Oct.27 | Mid-term exam | |
| Oct.28 | No lab | |
| Nov.3 | Lecture 8: Intro to GPs | |
| Nov.4 | Lab 7: GP regression | |
| Nov. 10 | Lecture 9: GP advanced topics (guest lecturer: A.Wilson) | |
| Nov. 11 | no lab, work on projects | |
| Nov.17 | Lecture 10. Deep learning for time series | |
| Nov.18 | Lab 8: RNNS | |
| Nov.24 | Lecture 11: Deep learning 1 | |
| Nov. 25 | no lab, work on projects | |
| Nov. 25 | Spectral methods 2 | |
| Dec. 1 | Lecture 12: Spectral methods | |
| Dec. 4 | Lab 9: Spectral methods | |
| Dec. 8 | Final projects presentation | Project reports due Dec.15 |
| Dec. 9 | No lab |
There is no required textbook. Assigned readings will come from freely-available online material.
- Time series analysis and its applications, by Shumway and Stoffer, 4th edition
- Pattern recognition and machine learning, Bishop
- Gaussian processes Rassmussen & Williams
- Review notes from Stanford's machine learning class
- Sam Roweis's probability review
- Carlos Ferndandez's notes on Statistics and Probability for Data Science DS-GA 1002
We expect you to try solving each problem set on your own. However, if stuck you should discuss things with other students in the class, subject to the following rules:
- Brainstorming and verbally discussing the problem with other colleagues ok, going together through possible solutions, but should not involve one student telling another a complete solution.
- Once you solve the homework, you must write up your solutions on your own.
- You must write down the names of any person with whom you discussed it. This will not affect your grade.
- Do not consult other people's solutions from similar courses.
- Credit should be explicitly given for any code you use that you did not write yourself.
- Violations result in a zero score on that assignment, and a notice to the DGS.
Penalties: 20% points off assignment for each extra day of delay.