Re: COMP: AI and mathematics: Proofs

From: Spike Jones (spike66@ibm.net)
Date: Sat Jul 22 2000 - 15:28:20 MDT


> At 06:33 PM 21/07/00 EDT, Greg wrote:
>
> >In a discussion in another forum, the question of whether computers have
> >"found" valid mathematical proofs has come up.

Sphere packing theorems have been demonstrated with computers.

Oh, and consider the abundance of primes. You know the abundance
of primes seems to go as the natural log of the number. That is, given
any number N, the probability N is prime is ln(N)/N.

There is no closed form proof that this relationship holds (that I
know of) but it has been demonstrated by computer to work for
very large values of N. The GIMPS project has shown the relationship
is still working waaaay out there.

Cool question Greg! spike



This archive was generated by hypermail 2b29 : Mon Oct 02 2000 - 17:35:06 MDT