[ml] Hi

[Brian Morris]

Hi Alexi

you might check out "reinforcement learning", which is a subtopic of ML more
like what most people think of as AI and relevant to robotics (not that the
rest of ML is irrelevant)

try googling, and for a short classic introduction, the slides from the last
chapter of Tom Mitchell's ML text:

http://www.cs.cmu.edu/afs/cs.cmu.edu/project/theo-20/www/mlbook/ch13.pdf

I may try and give a talk based on these slides some time (just finished
reading the book).

I also am learning Python, and want to learn Java, for these purposes. The
group here seems to be settling on R, which emphasizes (in Mitchell's
terminology) inductive (or statistical) learning. There is also Analytical
(deductive) learning and combinational approaches (cf Mitchell Chapter 11
and 12).
Bayesian learning is also pretty important in foundations of the field it
seems (I'm not an expert !)

Another sub-discipline is "unsupervised learning" (USL). The terminology
confused me a bit at first, they still use the term "training data" even
though targets are unspecified. This is like data mining or knowledge
discovery, ie

http://scikit-learn.sourceforge.net/tutorial.html#machine-learning-the-problem-setting

(note their definition of ML leaves out much of what I listed above … don't
ask me why)
(but also note sci-kits is developing implementations in USL -- if you are
interested in core development)


Brian


On Mon, Jul 18, 2011 at 11:08 AM, Alexi Kostibas <alexi at kostibas.com> wrote:

> Hey all,
>
> I'm interested in getting some hands-on experience with AI algorithms (GA,
> neural nets,) robotics (ROS?) and getting my head more in that space in
> general.
>
> I have varying levels of experience with C++, Python, Java, but no
> AI/robotics libraries as of yet.
>
> Basically, if there are any classes coming up, or projects needing a hand,
> I'm in! :)
>
> -Alexi
>
> _______________________________________________
> ml mailing list
> ml at lists.noisebridge.net
> https://www.noisebridge.net/mailman/listinfo/ml
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.noisebridge.net/pipermail/ml/attachments/20110718/1bc7984e/attachment-0003.html>