Scribd Logo
Upload

Welcome to Scribd!

  • 90 views

    DS-GA 3001.001 Special Topics in Data Science: Probabilistic Time Series Analysis

    Uploaded by

    Ougni Chakraborty
    This graduate level course covers probabilistic time series analysis methods like ARIMA, Kalman filtering, hidden Markov models, Gaussian processes, and recurrent neural networks. The course emphasizes generative models for time series and their inference and learning. It includes problem sets, a midterm exam, final project, and lab assignments. The final project involves applying algorithms to real data or improving existing methods in groups of 2-3 students. Lectures are on Mondays and labs are on Wednesdays. The course uses Python and provides video recordings of lectures.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd

    DS-GA 3001.001 Special Topics in Data Science: Probabilistic Time Series Analysis

    Uploaded by

    Ougni Chakraborty
    90 views3 pages
    This graduate level course covers probabilistic time series analysis methods like ARIMA, Kalman filtering, hidden Markov models, Gaussian processes, and recurrent neural networks. The course emphasizes generative models for time series and their inference and learning. It includes problem sets, a midterm exam, final project, and lab assignments. The final project involves applying algorithms to real data or improving existing methods in groups of 2-3 students. Lectures are on Mondays and labs are on Wednesdays. The course uses Python and provides video recordings of lectures.

    Original Description:

    ptsa

    Original Title

    syllabus

    Copyright

    © © All Rights Reserved

    Available Formats

    PDF, TXT or read online from Scribd

    Did you find this document useful?

    Is this content inappropriate?

    This graduate level course covers probabilistic time series analysis methods like ARIMA, Kalman filtering, hidden Markov models, Gaussian processes, and recurrent neural networks. The course emphasizes generative models for time series and their inference and learning. It includes problem sets, a midterm exam, final project, and lab assignments. The final project involves applying algorithms to real data or improving existing methods in groups of 2-3 students. Lectures are on Mondays and labs are on Wednesdays. The course uses Python and provides video recordings of lectures.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    90 views3 pages

    DS-GA 3001.001 Special Topics in Data Science: Probabilistic Time Series Analysis

    Uploaded by

    Ougni Chakraborty
    This graduate level course covers probabilistic time series analysis methods like ARIMA, Kalman filtering, hidden Markov models, Gaussian processes, and recurrent neural networks. The course emphasizes generative models for time series and their inference and learning. It includes problem sets, a midterm exam, final project, and lab assignments. The final project involves applying algorithms to real data or improving existing methods in groups of 2-3 students. Lectures are on Mondays and labs are on Wednesdays. The course uses Python and provides video recordings of lectures.

    Copyright:

    © All Rights Reserved
    Download as PDF, TXT or read online from Scribd
    Download as pdf or txt
    You are on page 1of 3

    DS-GA 3001.

    001 Special Topics in Data Science: Probabilistic Time Series


    Analysis

    Instructor: Cristina Savin


    TAs: Caroline Haimerl and Yiqiu (Artie) Shen

    Overview

    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 AR(I)MA, Kalman filtering,
    HMMs, RNNs, Gaussian Processes, and their application to several kinds of data.

    Note: information presented is tentative, syllabus may be subject to change as the


    course progresses.

    Course details
    Lecture (Section 008) 
    Mo 2.00pm - 3.40pm
    60 5th Avenue, Room 110

    Lab (Section 009)


    Wed 3:30pm-4:20pm
    60 5th Avenue, Room 110

    Enrollment cap: 64

    Web: https://github.com/savinteaching/pTSAfall2019

    Office hours
    CS: Mo 4:00-5:00pm. Location: 60 5th ave, 6th floor, room 608.
    CH/YS: TBD. Location: 60 5th ave, 6th floor, room 660.

    Grading
    problem sets (25%) +
    midterm exam (25%) +
    final project (25%) +
    lab(20%)+
    participation(5%)

    Participation: piazza, engagement during lectures, labs, and office hours


    Piazza
    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.

    Projects
    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.

    Video
    Video recordings of the lectures will be available via NYU Classes. Class attendance is
    still required.

    Date Lecture title Assignments

    Sept. 4 no lab

    Sept. 9 Lecture 1: Logistics. Introduction.


    Basic statistics for characterizing time
    series.

    Sept. 11 Recap basic Bayes, graphical models

    Sept. 16 Lecture 2: AR(I)MA Problem set 1 (due 2w later)

    Sept. 18 Lab 1: ARIMA

    Sept. 23 Lecture 3: LDS; Kalman filtering Projects initial discussion

    Sept. 25 Lab 2: Inference in LDS

    Sept. 30 Lecture 4: Particle filtering Problem set 2

    Oct. 2 Lab 3: LSD parameter learning

    Oct. 7 Lecture 5: Hidden Markov Models Project proposal due

    Oct. 9 Lab 4: Particle filtering

    Oct. 14 FALL RECESS

    Oct.15 (TUE!) Lecture 6: A unified view of linear Problem set 3


    models

    Oct.16 Lab 5: HMMs

    Oct.21 Introduction to Gaussian Processes

    Oct.23 No lab. Office hours


    Date Lecture title Assignments

    Oct.28 Mid-term exam

    Oct.30 No lab

    Nov. 4 GP advanced topics.Intro to RNNs Problem set 4

    Nov. 6 Lab GP

    Nov.11 Deep learning for time series

    Nov.13 RNNS lab

    Nov.18 Spectral methods 1 Problem set 5

    Nov. 20 Lab spectral methods

    Nov. 25 Spectral methods 2

    Nov. 27 Thanksgiving

    Dec. 2 Guest lecture: Joan Bello

    Dec. 4 No lab. Work on projects

    Dec. 9 Final projects presentation Project reports due Dec.15

    Dec. 11 No lab

    Bibliography
    There is no required textbook.
    Required reading will be provided on the course page in electronic format.
    We use python for all the lab work.

    Core material
    • Time series analysis and its applications, by Shumway and Stoffer, 4th edition
    • Pattern recognition and machine learning, Bishop
    • Gaussian processes, Rassmussen & Williams

    Useful extras
    • 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

    Policies
    Try to solve problems on your own first. If you get stuck, you can discuss homework
    questions with colleagues, but you need to write up the final solution individually. Any
    violation will be penalized with a zero score for he assignment and referred to the DGS.
    Credit should be explicitly given for any code you use that you did not write yourselves.
    Late submission penalties: 20% points off for each extra day of delay.

    You might also like

    pFad - Phonifier reborn

    Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

    Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


    Alternative Proxies:

    Alternative Proxy

    pFad Proxy

    pFad v3 Proxy

    pFad v4 Proxy