Machine Learning Theory

written by Theory Club

published February 07, 2022

Today Neil talked about ML theory (but better this time).

Rather than talk about random interesting results, tonight I’m going to try to present it in a more comprehensive way, with concepts like PAC, VC-dimensionality, Perceptron, Multi-armed Bandit, Experts, and Weighted Majority (ok, that was a lot, but I’ll try to get to as much of it as I can).

Slides from today can be found here.

I’ll try to add in the proofs later tonight in case anyone wants to reference them and here’s the link to Vempala’s version of the course from last semester: https://santoshv.github.io/MLTheory/MLcourse.html.