3
$\begingroup$

I'm a newbie trying to learn Boosting. The examples I found online are quite confusing. Is there a simple tutorial somewhere that explains Boosting with an example?

$\endgroup$
1
$\begingroup$

Recommended in comments:

A short note on boosting in the context of decision trees is provided in James et al. "An Introduction to Statistical Learning" p. 321-324. More detailed treatment is in Hastie et al. "The Elements of Statistical Learning" Chapter 10. – Richard Hardy

A working guide to boosted regression trees. Journal of Animal Ecology often used in introductory courses – charles

$\endgroup$
1
$\begingroup$

Machine Learning: Classification

This MOOC on Coursera by University of Washington has a very good and comprehensive explanation of boosting models in Week 5. They have specifically focused on Adaboost and have given a very good and easy to understand explanation of the model and the mathematics behind it. To get a glimpse of how the video is you can look at this pdf
I also found this article to be a very good intuitive explanation of XGBoost.

$\endgroup$
  • $\begingroup$ I really wish we'd stop focusing on AdaBoost. $\endgroup$ – Matthew Drury Dec 15 '18 at 22:27
  • $\begingroup$ @MatthewDrury: What would you propose to get the basics of a reference implementation based on your experience? AnyBoost (Mason et al.), LogitBoost (Friedman), something else? $\endgroup$ – usεr11852 says Reinstate Monic Dec 15 '18 at 23:57
  • $\begingroup$ Regression boosting to minimize least squares, here's an exposition I put together for a job interview: github.com/madrury/boosting-presentation/blob/master/latex/… $\endgroup$ – Matthew Drury Dec 17 '18 at 19:03
1
$\begingroup$

Chris' Bishop Pattern Recognition and Machine Learning has a full chapter ( Chapter 14) which addresses Bagging, boosting etc, def. worth a look, hope this helps!

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.