Machine Learning textbook with a Bayesian Perceptive

We had a discussion on the pymc_devs channel about what books to recommend to people wants to learn Machine Learning from a Bayesian perspective. I think it is worth to share:

To me, the classic Christopher Bishop is a must read:

@colcarroll and @fonnesbeck recommended Murphy along with Hastie & Tibshirani:


http://www.cs.ubc.ca/~murphyk/MLbook/


http://statweb.stanford.edu/~tibs/ElemStatLearn/

9 Likes

Bayesian Data Analysis, by Andrew Gelman, John Carlin, Hal Stern, David Dunson, Aki Vehtari, and Donald Rubin is another good read
http://www.stat.columbia.edu/~gelman/book/

2 Likes

The “puppy book” by John Kruschke is a good read, even for un-initiated. "Doing Bayesian Data Analysis"
https://sites.google.com/site/doingbayesiandataanalysis/

Murhpy is my go-to reference for most of ML

1 Like

I read the first few chapters of ‘Statistical Rethinking’ by McElreath and really like his style of writing.

In general I find that reading several authors on the same subject is very helpful to develop understanding.

https://www.crcpress.com/Statistical-Rethinking-A-Bayesian-Course-with-Examples-in-R-and-Stan/McElreath/p/book/9781482253443

1 Like

Yes, this book is really great. @aloctavodia has ported much of it to PyMC3.

1 Like