Message Boards Message Boards

[BOOK] Introduction to machine learning

Posted 3 years ago

WOLFRAM MATERIALS for the BOOK:

Etienne Bernard, Introduction to Machine Learning.

Wolfram Media: https://www.wolfram-media.com/products/introduction-to-machine-learning.html

Amazon: http://amzn.to/3oky6AC

ISBN: 978-1579550486

Code-only notebooks: https://wolfr.am/iml

enter image description here

Introduction

Hi everyone! For those who don't known me, I worked at Wolfram Research for 8 years (went pretty fast!) and led the development of the machine learning tools that are now part of the Wolfram Language (Classify, Predict, NetTrain, ...). It is now time for me to go on other adventures, but I couldn't leave "just like this", and decided to write a book called "Introduction to Machine Learning” http://amzn.to/3oky6AC to share my understanding of this field after these years of tool design and development. Writing this book has been quite of a journey, with some ups and downs, but it was overall pretty fun to play with the tools we built and figuring out the best ways to introduce machine learning. This book is intended for a general audience and has three goals:

  1. Explain what machine learning is.
  2. Teach how to practice machine learning.
  3. Give an understanding of how machine learning works.

It is written in a computational essay style (alternating text and simple computations - basically a long WL notebook), with lots of illustrations and examples. The code snippets are used to show how to practice machine learning, to illustrate concepts, and to complement - or even replace - mathematical formulations. It was quite nice to see how much math could be removed thanks to code (although, I often made the decision to explain concepts both with formulas and code snippets, to have the best of both worlds). Anyway, I encourage you to have a look at it, I included chapter 3 as a notebook in this post, I hope that you enjoy it or at least find it useful You can also check out the code-only notebooks https://wolfr.am/iml which contain both code snippets and illustrations code.

PS: I am starting a weekly #MLConcept post series on Twitter @etiennebcp to highlight the most important topics and start conversations. You are welcome to join in!

Sample chapter

POSTED BY: Etienne Bernard
5 Replies

enter image description here -- you have earned Featured Contributor Badge enter image description here Your exceptional post has been selected for our editorial column Staff Picks http://wolfr.am/StaffPicks and Your Profile is now distinguished by a Featured Contributor Badge and is displayed on the Featured Contributor Board. Thank you!

POSTED BY: EDITORIAL BOARD
Posted 3 years ago

I have a question about the book:

In the sample chapter, most of the material related to the use of built-in functions. Although the mushroom classification example does modify one layer of a supplied network.

Is there also material on the design of neural nets from scratch, including architecture and the rationale for the use of particular layer types?

POSTED BY: David Keith

Hi David, definitely, chapter 11 "Deep Learning Methods" (108 pages...) does exactly that. You can check this PDF sample where the beginning of chapter 11 is present https://t.co/pbYh0wJH0S.

Thanks, Etienne

POSTED BY: Etienne Bernard
Posted 3 years ago

Thank you, Etienne. This is just the book I’ve been looking for. I have it on order. Thank you for your contribution to Mathematica and best wishes for NuMind.

Kind regards, David

POSTED BY: David Keith

Hi Etienne, thank you for the book. I did learn a lot about neural Networks with Mathematica. Most of the code is running perfectly - it's really fun. Only a few lines and examples are not working. For example I can not get the decoder of the gpt2 network working.

Is there some Errata-Web-Page were I can find the corrections?

POSTED BY: Thomas Czinkota
Reply to this discussion
Community posts can be styled and formatted using the Markdown syntax.
Reply Preview
Attachments
Remove
or Discard

Group Abstract Group Abstract