Re: COMP: AI and mathematics: Proofs

From: Damien Broderick (d.broderick@english.unimelb.edu.au)
Date: Sat Jul 22 2000 - 01:01:41 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.

Newell and Simon's very first program, Logical Theorist, proved (from first
principles) theorems from the Whitehead/Russell PRINCIPIA. In 1956, running
on Rand's Johnniac, it produced the complete proof of theorem 2.01; indeed,
it proved 38 of the first 52 theorems in chapter two, mostly in under a
minute each. You'll recall that Russell nearly killed himself doing this work.

Damien Broderick



This archive was generated by hypermail 2b29 : Mon Oct 02 2000 - 17:35:05 MDT