[ml] [Noisebridge-discuss] Probability study group
josh at joshisanerd.com
Mon Nov 1 22:28:22 PDT 2010
FWIW, I'm about halfway through a Monte Carlo implementation of the
questions in "50 Challenging Problems in Probability," which is a great
resource for Mastery Questions.
http://blog.appliedplatonics.com/tag/fifty-problems/ is an index of them.
I'm now pretty firmly in the camp that believes nuanced probability
formulations are mostly obsolete, and it's sufficient to do regressions on
top of a bunch of Monte Carlo simulations. Some problems are intractable
(due to combinatorial explosion, etc), but computers are just too fast for
me to spend a lot of time getting hung up in the ways one can simplify (or
symbolically approximate) exact solutions.
(Don't tell the me of five years ago, he'd be kind of pissed that I went
over to the SVMs-suffice camp.)
On Mon, Nov 1, 2010 at 10:12 PM, Ben Weisburd <ben.weisburd at gmail.com>wrote:
> Great to see there's some interest.
> Kai - thanks for setting up the wiki. (
> I just put up a proposed meeting format on there and also a set of chapter
> 1 problems which we can hopefully go over at the 1st meeting.
> To figure out the meeting time, can everybody please put in what time works
> for you at:
> Look forward to meeting you all,
> Ps. Does somebody know how to set up a mailing list?
> On Mon, Nov 1, 2010 at 5:25 PM, Kai Chang <kai.salmon.chang at gmail.com>wrote:
>> I'd be down for teaching a chapter.
>> On Mon, Nov 1, 2010 at 3:48 PM, Mike Schachter <mike at mindmech.com> wrote:
>>> Hey Ben,
>>> I'd be interested in meeting up but I'm usually pressed for time. I
>>> cc'ed the machine learning group - probability is pretty important
>>> for them and surely some ML'ers are interested!
>>> On Mon, Nov 1, 2010 at 3:44 PM, Ben Weisburd <ben.weisburd at gmail.com>wrote:
>>>> Is anyone interested in joining up to learn probability over the next
>>>> 1.5 months or so?
>>>> There is a great text book that's available free online: Fundamentals
>>>> of Applied Probability Theory by Al Drake<http://ocw.mit.edu/courses/electrical-engineering-and-computer-science/6-041-probabilistic-systems-analysis-and-applied-probability-spring-2006/related-resources/>
>>>> The author taught probability at MIT for many years.
>>>> Ideally, we would meet once a week and go over 1 chapter. Depending on
>>>> people's interest, we could either have one person teach the chapter to
>>>> everyone else during the meeting, or we could each read the chapter at home
>>>> and do 10-or-so of the recommended problems. Then, we could spend the
>>>> meeting discussing what we found interesting/confusing and comparing answers
>>>> to the problems.
>>>> Noisebridge-discuss mailing list
>>>> Noisebridge-discuss at lists.noisebridge.net
>>> ml mailing list
>>> ml at lists.noisebridge.net
> ml mailing list
> ml at lists.noisebridge.net
Josh Myer 650.248.3796
josh at joshisanerd.com
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the ml