ML module Week 2 Topic 1: Intro to ML #47
Description
Broad overview of ML landscape. What is it, why is it, types of ML, etc.
Explain what machine learning is and what kinds of problems it solves.
This is a new "chapter" in the CTD trajectory. Many students haven't been exposed to ML before, so we'll need a nice intro here with a why. We'll want to explain the "classic" diagram with AI-> ML -> Deep learning etc and how we are going to go through a section of this (https://miro.medium.com/v2/resize:fit:1100/format:webp/0*IWANMRSPfUJMNrai.png)
We will go through classical ML, deep learning (and in the next module of the course, LLMs).
We should stress that not all problems require neural networks, often simple problems can be solved with simpler techniques. We'll walk through some of these first (and techniques for evaluating ML algorithms are easier to learn on the simpler methods, so that's a great place to start).
Remember this isn't about just conveying information to the students, but motivation. There are tons of great resources out there about ML, so lean on those and provide links to those resources. We aren't reinventing the wheel.
Distinguish types of machine learning problems.
- supervised vs. unsupervised learning (clustering)
- regression vs classification
The main demarcations. We will focus on supervised, and hands on with regression and classification. But in the following section on scikit-learn we will do on eexample of clustering...