Book club “Hands-On Machine Learning with R” #4 - Regularised Regression

Want to know more about regularisation? Join us for the FOURTH meetup for the online boookclub discussing ‘Hands-On Machine Learning with R’ by Bradley Boehmke and Brandon Greenwell. On Monday October 24, we discuss Chapter 6, ‘Regularised Regression’.

We’re delighted to announce that this topic will be presented by Marianna, who is from the R-Ladies community! Marianna is a Ph.D. student in economics, she will be defending her thesis this November after which she will start a post-doc at the University of Barcelona. About R, Marianna says: “I got acquainted with R a couple of years ago, to me first the data-wrangling and data visualization in R was very appealing. Currently, I am using it also for regression analyses, and I am doing my papers in Rmarkdown.” About the topic of this bookclub edition, regularised regression, Marianna says: “I wanted to do a topic that would allow me to learn new things and I hope I can apply it in the future.”

When your model contains a lot of predictors, it is likely that your model assumptions, like “no multicollinearity”, will no longer hold. Regularisation is a way to deal with this problem by shrinking the model coefficients. During the meeting we will go into more detail on this topic and talk about different kinds of regularised regression.

For this boook club we - the Den Bosch and Utrecht chapters of R Ladies - are joining forces. We will alternate hosting the meetups. In +- 10 meetings (maybe more :) ) , we will work through all the chapters in the book together. We are doing this boookclub, because we like to learn together. It is not the case presenters are the experts on the matter. To make this a group effort, the idea is that all participants take turns in presenting chapters, as for example Marianna will be presenting chapter 6! If you’re not comfortable presenting, just listening is also okay.

If you cannot make it this time, feel free to join in on one of the subsequent meetings.

For who:

• People who want to take their first steps into machine learning
• Some prior knowledge on R programming will be helpful. As the authors of the book themselves mention: “This book is not meant to be an introduction to R or to programming in general; as we assume the reader has familiarity with the R language to include defining functions, managing R objects, controlling the flow of a program, and other basic tasks.”

What to expect:

• Presentations summarising the key points on each topic of the book
• Putting the theory to practice by going through exercises together
• Room to discuss questions and other thoughts

Whereabouts:

• Meet every 2 weeks via Zoom. The password will be sent to you via e-mail before the meeting.
• We meet every second Monday, from 7 - 8.30 pm
• It is not necessary to read the chapters in advance, but feel free to do so.

R-Ladies is dedicated to providing a harassment-free experience for everyone. We do not tolerate harassment of participants in any form. See the code of conduct.

The book can be found at https://bradleyboehmke.github.io/HOML/. It is also available in print. For the slides of our previous meetings, see https://github.com/rladiesnl/book_club_handsonML.