Because unless I know otherwise, I use simplicity-preferring prior.
Yes, absolutely. Occam's razor, etc.
However, when you have a local effect that is mediated at a global level,
then you *NEED* to have that as part of your computational model -- and
neural networks simply don't take such things into account.
It is very easy, as J. Rogers
noted, to scramble a simple object so that you won't ever guess that
it's simple based just on how it looks on the surface.
True. But it's even easier to simplify something that is only moderately
complicated so far that it will never be able to reproduce real-world
behavior.
At any rate, as Richard points out, y'all are so far from reality that
arguing with you is not a wise use of time. Do what you want to do. The
proof will be in how far you get.
----- Original Message -----
From: "Vladimir Nesov" <[EMAIL PROTECTED]>
To: <agi@v2.listbox.com>
Sent: Monday, June 02, 2008 12:01 PM
Subject: Re: [agi] Ideological Interactions Need to be Studied
On Mon, Jun 2, 2008 at 6:27 AM, Mark Waser <[EMAIL PROTECTED]> wrote:
But, why "SHOULD" there be a *simple* model that produces the same
capabilities?
What if the brain truly is a conglomeration of many complex interacting
pieces?
Because unless I know otherwise, I use simplicity-preferring prior.
Biological complexity is hardly evidence that the refined
computational model is also complex. It is very easy, as J. Rogers
noted, to scramble a simple object so that you won't ever guess that
it's simple based just on how it looks on the surface.
--
Vladimir Nesov
[EMAIL PROTECTED]
-------------------------------------------
agi
Archives: http://www.listbox.com/member/archive/303/=now
RSS Feed: http://www.listbox.com/member/archive/rss/303/
Modify Your Subscription:
http://www.listbox.com/member/?&;
Powered by Listbox: http://www.listbox.com
-------------------------------------------
agi
Archives: http://www.listbox.com/member/archive/303/=now
RSS Feed: http://www.listbox.com/member/archive/rss/303/
Modify Your Subscription:
http://www.listbox.com/member/?member_id=8660244&id_secret=103754539-40ed26
Powered by Listbox: http://www.listbox.com
