Course Content

This page includes all the class content for the course thus far. We will update this page with lecture materials, lab materials, and readings as the class goes on.

The class is roughly split into two parts. The “classical ML” part will concern foundational techniques and ideas that underlie the subfield of machine learning from a unifying statistical learning framework. Sam will be the lecturer in this part. The “modern ML” part will concern deep learning and techniques that have gained traction in the past decade because of the proliferation of data and computational power. Nick will be the lecturer in this part.

This is a tentative schedule and is subject to change.

Lecture Recordings. Recordings of the lectures are available to watch on Brightspace. Just navigate to the Brightspace page and then click Zoom.

Math Review. As stated in the prerequisites, students are expected to have a solid understanding of linear algebra, probability and statistics, and multivariable calculus for this course. In addition to the resources listed on the homepage of the site, here are several shorter notes to get you started if you need to review material:

Sam also designed a course the past two summers at Columbia meant to give students a deeper understanding of these prerequisites (given that they have already taken them and would like to progress to graduate-level machine learning). The entire course could be found at this page: Math for ML, and lecture videos can be found in “Video Recordings” on this page.

Math Review Videos. We will also occasionally post some Math Review Videos after that week’s lab depending on lab feedback on what students found confusing about the material in that week. Familiarity with all the prerequisites in this class can definitely feel daunting at times, so the purpose of these review videos are to re-introduce some prereq concepts or lecture derivations at a slower pace. These are a totally optional extra resource. Hopefully this helps you digest or page in you may have been rusty on!

Part 1: “Classical” ML (Weeks 1–11)

Week 1: Introduction & Basic Statistical Learning Framework

Week 2: Optimization & Gradient Descent

Week 3: Regularization & Loss Functions

Feb 3
Lecture Lecture 3: Regularization & Loss Functions
ESL Ch. 3 or ISL Ch. 6.1-6.2 for a lighter intro; PPA pg. 92-98, PPA pg. 100-106 (for a more “modern” treatment of regularization), PPA pg. 122-125 (if you’re really interested in “modern treatment” of regularization)
Feb 5
Lab

Week 4: Linear Classification, Convex Optimization, & SVM

Feb 10
Lecture
Feb 12
Lab

Week 5: President's Day (Class/Lab Cancelled)

Feb 17
No class this week!
Feb 19
No lab this week!

Week 6: Features & Kernels

Feb 24
Lecture
Feb 26
Lab

Week 7: MLE & Conditional Probability Models

Mar 3
Lecture
Mar 5
Lab

Week 8: Midterm

Mar 10
Midterm
Mar 12
No lab this week!

Week 9: Spring Break (Mar 16-20)

Mar 17
No class this week!
Mar 19
No lab this week!

Week 10: Decision Trees, Bagging, & Random Forests

Mar 24
Lecture
Mar 26
Lab

Week 11: Boosting

Mar 31
Lecture
Apr 2
Lab

Part 2: Modern ML (Weeks 12–16)

Week 12: Learnable Features → Deep Neural Networks

Apr 7
Lecture
Apr 9
Lab

Week 13: Structured Neural Networks

Apr 14
Lecture
Apr 16
Lab

Week 14: Generative Models

Apr 21
Lecture
Apr 23
Lab

Week 15: Reinforcement Learning

Apr 28
Lecture
Apr 30
Lab

Week 16: Language Models

May 5
Lecture
May 7
End of semester, no lab this week!