To answer your questions:
Right now the most machines we have used for a single NM installation is 4
However, scaling up to many machines is NOT our biggest issue by any means...
In 2000, we ran our Webmind AI Engine system (with a conceptually
similar distributed processing infrastructure) on a network of 30+
quad processor PC's ...
This kind of infrastructural stuff is not the obstacle to AGI, the
obstacle is in getting the learning/memory algorithms right.... I
think the NM design contains a workable solution to this problem but
there remain plenty of details to resolve via experimentation...
NM is not at this point designing itself, no -- it is still at the
"infantile" stage...
ben
On Wed, 31 May 2006 11:09:44 -500, [EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
What is the largest test to date of Novamate on a distributed network of
machines?
Is Novamate designing itself?
Dan Goe
----------------------------------------------------
From : Ben Goertzel <[EMAIL PROTECTED]>
To : agi@v2.listbox.com
Subject : Re: [agi] Data there vs data not there, Limits to storage?
Date : Wed, 31 May 2006 11:51:45 -0400
> Novamente can run on a distributed network of machines, using both RAM
> and disk for memory; hence in principle it has no "limits of storage"
> except those imposed by cost...
>
> Of course there are limitations in terms of RAM access time and
> distributed processing overhead, but these are the familiar ones and
> not specific to Novamente
>
> ben
>
-------
To unsubscribe, change your address, or temporarily deactivate your
subscription,
please go to http://v2.listbox.com/member/[EMAIL PROTECTED]
-------
To unsubscribe, change your address, or temporarily deactivate your subscription,
please go to http://v2.listbox.com/member/[EMAIL PROTECTED]
