Computer Chess Club Archives


Search

Terms

Messages

Subject: Re: Turing tested at last?

Author: Ricardo Gibert

Date: 09:45:53 11/19/00

Go up one level in this thread


On November 19, 2000 at 11:42:04, Mike S. wrote:

>On November 19, 2000 at 11:05:41, Ricardo Gibert wrote:
>
>>On November 19, 2000 at 10:25:21, Hermano Ecuadoriano wrote:
>>
>>>There has been some discussion here about holding some
>>>"exhibition" variations on the Turing test.
>>>This must be done eventually.
>>>If successful, it would be epoch-making.
>>>I think the year 2001 is fantastically apropo,
>>>promotionally speaking!
>>
>>I don't think any of the top computer programs can come close to passing a
>>turing test. (...)
>>
>>I would not be surprised if the CCC membership could easily devise a test
>>consisting of 10 positions, which virtually any strong human would solve 10 of
>>10 and the top programs would solve 0 of 10.
>
>Probably possible, but this would not be a Turing test (at least not compliant
>to the definition I remember). A it was just discussed below, referring to an
>announcement of Ed Schröder regarding computer games against GM v.d.Wiel, this
>would be a test by watching a game (live or afterwards) without knowing which of
>the players if the human and which one is the computer.

Your take on what the turing test consists of raised an eyebrow. Now I have to
break out a book I have on AI called "Artificial Intelligence: a modern
approach" by Stuart Russell & Peter Novig. This is how they define it:

"The Turing Test, proposed by Alan Turing (1950), was designed to provide a
satisfactory operational definition of intelligence. Turing defined intelligent
behavior as the ability to achieve human-level performance in all cognitive
tasks, sufficient to fool an interrogator. Roughly speaking, the test he
proposed is that the computer should be interrogated by a human via a teletype,
and passes the test if the interrogator cannot tell if there is a computer or a
human at the other end."

In other words, as the interrogator I would easily flunk todays top computer
programs with a 10 position test. What the test consists of is clearly the
choice of the interrogator.

>
>It would indeed be interesting to see, if chess experts can determine this...
>Did computerchess catch up with science fiction in this respect? As Hermano
>indicates by *2001*, playing chess like HAL is different than "just" playing
>very strong like i.e. Deep Blue:
>
>http://mitpress.mit.edu/e-books/Hal/chap5.java/five1.html
>
>Regards,
>M.Scheidl
>
>
>P.S. If someone wants to try it:
>http://x56.deja.com/=dnc/getdoc.xp?AN=657997698&CONTEXT=974605131.1065680911&hitnum=0



This page took 0 seconds to execute

Last modified: Thu, 15 Apr 21 08:11:13 -0700

Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.