How about parallel computing with finance
Horatio B. Bogbindero
wyy at cersa.admu.edu.ph
Wed Dec 6 05:39:15 PST 2000
i would just like to know about building neural networks in clusters.
i am not into neural network but some people here in the university
maybe interested. however, we do not know where to start. i would to know
where i can get some sample NN code. maybe for something trivial.
On Wed, 6 Dec 2000, Robert G. Brown wrote:
> On Wed, 6 Dec 2000, liuxg wrote:
> > Hi, all£¡
> > Now using parallel computing in the solutiong of
> > computionally-intensive probolems has been widely accepted
> > in physics,chemistry,biology,engineering etc. I'm inerested
> > in its application in economics and finance based linux cluster.
> > I know some peoples are busying in it , such as research in
> > www.doc.ic.ac.uk, but I need more info abou the field . Does anybody
> > can help me ?
> Market Driven (a small startup I'm associated with) has a neural engine
> that is parallelized that is used to build predictive models on a small
> beowulf in house. It also has the expertise to build more advanced
> parallelized tools if/when the demands of the market require them.
> The engine can be applied to all sorts of things in business and finance
> (and medicine and science and...). Our primary applications to date
> have been in scientific marketing, but we've also done a pilot project
> on predicting outcomes for (possible) breast cancer patients based on
> data from physical and radiological exams, for example. I wrote the
> engine itself, so we have complete control and the ability to customize
> it and tune it for any particular application that might arise, or to
> custom integrate it with any given customer's application that requires
> a powerful modeling engine. My primary partner is a database guru and
> our employees are skilled in database and web stuff both (and we can all
> cut code), so we can do custom integrations and more with comparative
> Modeling is just one place where beowulf technology is useful in
> business, but it is a very important one. In some cases models can be
> built over weeks and applied over months; in others (for example, to
> drive an ongoing direct phone sales campaign) the turnaround based on
> real time feedback needs to be hours or at most overnight. In this case
> being able to parallelize means that an analysis that might take days
> (we build >>serious<< networks:-) is cut down to hours.
> Hope this helps,
> Robert G. Brown http://www.phy.duke.edu/~rgb/
> Duke University Dept. of Physics, Box 90305
> Durham, N.C. 27708-0305
> Phone: 1-919-660-2567 Fax: 919-660-2525 email:rgb at phy.duke.edu
> Beowulf mailing list
> Beowulf at beowulf.org
william.s.yu at ieee.org
"I am, therefore I am."
More information about the Beowulf