[pymvpa] Introduction to Machine Learning and SVMs

[J.A. Etzel]

Not written as a tutorial per se, but I've found the discussion in this 
paper useful for helping people understand how the machine learning 
approach works in a general/conceptual sense:

Breiman, L., 2001a. Statistical modeling: the two cultures. Stat. Sci.
16, 199–215.

Jo


On 5/16/2011 11:09 AM, Yaroslav Halchenko wrote:
> IIRC the best video describing SVMs (with math though) is
> http://videolectures.net/mlss06tw_lin_svm/ from Support Vector
> Machines author:Chih-Jen Lin, National Taiwan University
>
> who is an author of libsvm
>
> as for "less math" -- need to think about it... and it would depend
> on what aspects of SVM you want them to accent in comparison to other
> methods
>
> On Mon, 16 May 2011, Thorsten Kranz wrote:
>
>> Hi all,
>
>> I have a question, maybe you have a quick reply (to a non-trivial
>> question though...).
>
>> Here in my lab, some colleagues without too much knowledge in
>> mathematics would like to learn (and understand) some basics of
>> machine learning and SVMs in particular, so we'll have a little
>> methods-seminar soon. I will try to explain it to them, but it
>> would be nice if I could send them some kind of tutorial-paper or
>> book-chapter they could read before that.
>
>> Do you have any proposal for that? I know of the Hastie et al.
>> book online, but maybe "less mathematics" would fit better to
>> (some) of my colleagues.
>
>> Thanks in advance, greetings,
>
>> Thorsten
>
>> _______________________________________________ Pkg-ExpPsy-PyMVPA
>> mailing list Pkg-ExpPsy-PyMVPA at lists.alioth.debian.org
>> http://lists.alioth.debian.org/mailman/listinfo/pkg-exppsy-pymvpa
>
>