Data Science

Logistics

Location Virtual, with potential on-campus meetings if Covid situation improves
Time Mondays 9:30 AM on Zoom
Mailing List https://piazza.com/umd/fall2020/inst808/home
Suggested Text Natural Language Processing with PyTorch
Homework Template https://github.com/ezubaric/methods-hw
Syllabus https://docs.google.com/document/d/13jRziPi2jWEmS19Ll8rfD1Y7heztukb3uxM9sj-cr8E/edit?usp=sharing
Homework Turn-in https://umd.instructure.com/courses/1288606

People

Professor

Jordan Boyd-Graber

Schedule

How to read this table:

  1. Do the reading under the corresponding date.
  2. Homeworks are due at 11:55 PM (Eastern) the day listed on the schedule.
Date Topic Assignment Due Materials
Mon 31. Aug Course Introduction, Python Review, Math Review
Videos: Resources:
Mon 14. Sept Math Review: Distributions, Statistics, Vectors, Matrices, Derivative and Optimization [Slides]
Videos: Reading:
Mon 21. Sept Representations [In-Class Exercise]
Reading:
  • First half of Chapter 1
Videos:
Wed 23. Sep Homework 0 Due: Python / Math Warmup [Github]
Mon 28. Sept Hypothesis Testing [Slides]
Videos: Reading:
Mon 5. Oct Linear and Logistic Regression [Slides]
Videos:
  • Linear regression
  • Logistic regression for classification
  • Logistic regression objective function
  • SGD for logistic regression
  • Feature engineering and normalization
Mon 5. Oct Introducing PyTorch
Reading:
  • Second Half of Chapter 1
Videos:
Wed 14. Oct Homework 1 Due: Logistic Regression [Github]
Mon 12. Oct Clustering and Topic Models
Mon 19. Oct Feed Forward Networks
Mon 26. Oct RNNs and LSTMs
Mon 2. Nov Potpurri: Loading Data, Normalization
Mon 9. Nov Exam
Mon 16. Nov Visualization
Mon 23. Nov Sequence to Sequence
Mon 30. Nov Debugging and Understanding Models
Mon 7. Dec Transformers and BERT