Chapter 1 About

This book is being written as part of a final project for 797ML at UMass Amherst, spring 2022. It contains a simple reference and breakdown for a couple of dozen core methods used in machine learning.

The intent is twofold:

  1. Serve as a reference for the basics of the material covered in the class, using language and examples that are as simple as possible to explain the core concepts and how to do them;
  2. Force myself to learn these techniques better by carrying out the above.

The main purpose of this work is to be simple, not to be comprehensive. We won’t cover every facet of every technique, or every possible permutations of outcomes. The goal is to simply express the broad strokes and core concepts in a way that can be easily remembered, and to serve as a jumping-off point when more information is needed.

1.1 Authoring guidelines

The goal is to have no more than a few short paragraphs for each section, and to keep each explanation of the meanings of outcome variables to one sentence each.

Each chapter will have a TL;DR section at the top with one-sentence answers to following questions:

  • What it does (answer the question starting with “It…”)
  • When to do it
  • How to do it
  • How to assess it

and it will be possible to skim through the book, just surfing the TL;DR at the top of each chapter, and get a reasonable overview without reading the rest. The rest of each chapter will deliver more fleshed-out, but short, answers to the same questions, covering the basics and the most general concepts.

1.3 Resources

A lot of the material from this work is from the class textbook, James et al. (2021). I also find UNC geneticist Josh Starmer’s StatQuest video series on YouTube to be immensely helpful for simple explanations of statistics and machine learning concepts. His website statquest.org has many additional useful resources.

References

James, Gareth, Daniela Witten, Trevor Hastie, and Robert Tibshirani. 2021. An Introduction to Statistical Learning. 2nd ed. Springer. https://link.springer.com/book/10.1007/978-1-0716-1418-1.