26 ms·

Free book on Bayesian machine learning by David Barber

146 points by markerdmann 15y ago

mjw 15y ago
Having taken some courses from David and others at UCL recently, I'm a big fan of this. The Bayesian modelling perspective I think is very useful if you're interested in machine learning as more than just a collection of clever algorithms and optimisation techniques to throw at a problem and see what sticks. (Not that this isn't useful sometimes...) It provided a lot of motivation and unifying intuition for me anyway. The elegance of having a nice statistical model doesn't come for free though, there are some tricky computational issues associated with inference in many Bayesian models. The book covers them in some depth and seems quite a useful reference into the state of the art as well as a nice introduction to the area.
- the_cat_kittles 15y ago
  W/r/t computing gnarly integrals, interested parties might appreciate pymc, a python package that implements markov chain monte carlo methods to estimate them
reader5000 15y ago
Assuming this is legitimately released (seems to be), authors who write and release these books for free are heroes for those of us not currently undergrads at Stanford etc.
- StavrosK 15y ago
  It is, I've taken this class and David Barber is releasing this for free. That's the best thing I can say about that class, though.
parrisj 15y ago
Just a couple of notes 1) The linked version is out of date here the most current version as of Nov 2011 http://web4.cs.ucl.ac.uk/staff/D.Barber/textbook/211111.pdf http://web4.cs.ucl.ac.uk/staff/D.Barber/textbook/211111.pdf 2) @reader5000 It's legit he links to it from his homepage http://web4.cs.ucl.ac.uk/staff/D.Barber/pmwiki/pmwiki.php?n=Main.Textbook http://web4.cs.ucl.ac.uk/staff/D.Barber/pmwiki/pmwiki.php?n=...
mindcrime 15y ago
On a related note, there are tons of gems like this out there, and there are a handful of awesome sub-reddits dedicated to keeping lists of them: http://csbooks.reddit.com http://csbooks.reddit.com http://physicsbooks.reddit.com http://physicsbooks.reddit.com http://mathbooks.reddit.com http://mathbooks.reddit.com http://econbooks.reddit.com http://econbooks.reddit.com http://eebooks.reddit.com/ http://eebooks.reddit.com/ etc.
solusglobus 15y ago
The latest version of the book can be found at the author's page:- [1] http://web4.cs.ucl.ac.uk/staff/D.Barber/pmwiki/pmwiki.php?n=Main.Textbook http://web4.cs.ucl.ac.uk/staff/D.Barber/pmwiki/pmwiki.php?n=... [2] Direct link: http://web4.cs.ucl.ac.uk/staff/D.Barber/textbook/211111.pdf http://web4.cs.ucl.ac.uk/staff/D.Barber/textbook/211111.pdf
jwr 15y ago
This is one of the best resources for learning about Bayesian ML methods if you need a gentle introduction. The only other book I found which was similarly clear and well thought-out is Christopher Bishop's "Pattern Recognition and Machine Learning".
- EmilStenstrom 15y ago
  I'm not sure I would call 600 pages of heavy math a "gentle introduction".
  - grad_ml 15y ago
    Sorry to differ , but Barber's book is certainly gentler then Bishop. But Bishop book is amazing , full of amazing insights !
laacz 15y ago
I'm sure, that because of TeX and stuff, it's popular among lots of people to publish their free (or non-free) e-books as PDFs. Still, because of small screen reading devices, it would be great if they could publish an epub also. Or source. Or anything convertible to epub/mobi.

146 points by markerdmann 15y ago

mjw 15y ago

Having taken some courses from David and others at UCL recently, I'm a big fan of this. The Bayesian modelling perspective I think is very useful if you're interested in machine learning as more than just a collection of clever algorithms and optimisation techniques to throw at a problem and see what sticks. (Not that this isn't useful sometimes...) It provided a lot of motivation and unifying intuition for me anyway. The elegance of having a nice statistical model doesn't come for free though, there are some tricky computational issues associated with inference in many Bayesian models. The book covers them in some depth and seems quite a useful reference into the state of the art as well as a nice introduction to the area.

the_cat_kittles 15y ago
W/r/t computing gnarly integrals, interested parties might appreciate pymc, a python package that implements markov chain monte carlo methods to estimate them

reader5000 15y ago

Assuming this is legitimately released (seems to be), authors who write and release these books for free are heroes for those of us not currently undergrads at Stanford etc.

StavrosK 15y ago
It is, I've taken this class and David Barber is releasing this for free. That's the best thing I can say about that class, though.

parrisj 15y ago

Just a couple of notes 1) The linked version is out of date here the most current version as of Nov 2011 http://web4.cs.ucl.ac.uk/staff/D.Barber/textbook/211111.pdf http://web4.cs.ucl.ac.uk/staff/D.Barber/textbook/211111.pdf 2) @reader5000 It's legit he links to it from his homepage http://web4.cs.ucl.ac.uk/staff/D.Barber/pmwiki/pmwiki.php?n=Main.Textbook http://web4.cs.ucl.ac.uk/staff/D.Barber/pmwiki/pmwiki.php?n=...

mindcrime 15y ago

On a related note, there are tons of gems like this out there, and there are a handful of awesome sub-reddits dedicated to keeping lists of them: http://csbooks.reddit.com http://csbooks.reddit.com http://physicsbooks.reddit.com http://physicsbooks.reddit.com http://mathbooks.reddit.com http://mathbooks.reddit.com http://econbooks.reddit.com http://econbooks.reddit.com http://eebooks.reddit.com/ http://eebooks.reddit.com/ etc.

solusglobus 15y ago

The latest version of the book can be found at the author's page:- [1] http://web4.cs.ucl.ac.uk/staff/D.Barber/pmwiki/pmwiki.php?n=Main.Textbook http://web4.cs.ucl.ac.uk/staff/D.Barber/pmwiki/pmwiki.php?n=... [2] Direct link: http://web4.cs.ucl.ac.uk/staff/D.Barber/textbook/211111.pdf http://web4.cs.ucl.ac.uk/staff/D.Barber/textbook/211111.pdf

jwr 15y ago

This is one of the best resources for learning about Bayesian ML methods if you need a gentle introduction. The only other book I found which was similarly clear and well thought-out is Christopher Bishop's "Pattern Recognition and Machine Learning".

EmilStenstrom 15y ago
I'm not sure I would call 600 pages of heavy math a "gentle introduction".
- grad_ml 15y ago
  Sorry to differ , but Barber's book is certainly gentler then Bishop. But Bishop book is amazing , full of amazing insights !

laacz 15y ago

I'm sure, that because of TeX and stuff, it's popular among lots of people to publish their free (or non-free) e-books as PDFs. Still, because of small screen reading devices, it would be great if they could publish an epub also. Or source. Or anything convertible to epub/mobi.