"Georg Bauhaus" <rm.tsoh.plus-bug.bauhaus@[EMAIL PROTECTED]
> wrote in
message news:4805107d$0$636$9b4e6d93@[EMAIL PROTECTED]
> roderick.chapman@[EMAIL PROTECTED]
wrote:
>> On Apr 14, 3:59 pm, roderick.chap...@[EMAIL PROTECTED]
wrote:
>>> Today's new issue of GCN
(athttp://www.gcn.com/print/27_8/46116-1.html)
>>
>> Interestingly, this article as just been picked up by SlashDot...
>
> As might have been expected, the article offers a lot to
> foster the stereotypes about the expensive language that
> drives weapons home; Ada is nothing more than Pascal, and
> not a word about the fundamental type system and concurrency
> until very late in the game. So the least interesting remarks
> make it to the top. Oh well...
Oh well, indeed. Leemon Baird is a good guy, but they are talking about
my
student Tyler Hallmark (at West Point, not the Air Force Academy)
regarding
the Connect 4 re****t. Leemon's presentation - at least the one I saw -
was
about implementing a neural net and getting the same performance as C.
:-S
Tyler did some nice work for an undergrad effort.
ABSTRACT
Parallel evolution of game evaluation functions in ada
This is an Ada experience re****t, where we conclude that Ada tasking and
distributed processing facilities make it a good research tool for
experimentation with algorithms that might eventually need multiple
processors. We implemented a genetic algorithm in Ada to create effective
computer players for Connect4. Key to our success was employing Ada
tasking
and ALRM Annex E Distributed computing to harness a symmetric
multiproces-sor and a distributed machine with very few code changes. Easy
extension of an original single-task code to multi-tasking and distributed
variants-even though extension was not planned in advance-was essential to
timely completion. Using either the parallel or distributed
implementation,
about 150 processor hours were sufficient to evolve players that neither
the
GNU "Four-in-a-Row" Expert player nor the author could defeat. This
algorithm relies on human expertise to restrict the genetic search space.
Work is in progress on a new algorithm with near-zero encoded knowledge,
which will run on 220 distributed nodes within the same distributed
computing framework.
|
