Author: Graham Laight
Date: 02:59:37 01/11/99
Go up one level in this thread
On January 10, 1999 at 23:02:33, Oliver Y. wrote:
>On January 10, 1999 at 19:01:17, Graham Laight wrote:
>
>>On January 10, 1999 at 09:04:50, Oliver Y. wrote:
>>
{snip}
>>>Sorry, I guess I should just post the games already...but there's really no way
>>>in the next 20 years you'll see programs finding effective swindles against
>>>humans...
>>>
>>>That would be a true sign of Artificial Intelligence, IMMHO.
>>
>>I disagree with Oliver.
>>
>>Winning a lost position does not require special intelligence - rather it
>>requires a mistake by the opponent.
>
>Firstly, you are obfuscating the matter by using the term special.
Guilty as charged.
I often correct other people's English, so I must accept the criticisms when
they are levelled at me.
>Secondly, it most certainly requires intelligence to win a lost position.
>Your last clause is redundant--show me a single game where you can prove that
>no mistake has been made.
>
>Allow me to simplify. Can you tell if this post is composed by a computer or
>a human?
I assume you're human because computers don't write stuff like this yet.
However, I'm sure you'll agree that the ability to write stuff like this does
not mean that you're any good at chess.
>Assume you are watching a game where the contestant who is currently losing is
>hidden, so that you do not know whether it is a computer or a human.
>
>The opponent that is visible to you is not perfect, so you can have either a
>human or a computer in that role.
>
>The hidden player makes moves in a manner that fools the visible imperfect
>player into a loss or a draw, from a totally lost position--so a successful
>swindle(s) has occurred.
>
>Surely you would conclude that there was evidence of intelligence displayed by
>this hidden opponent, would you not?
>
>Moreover, if the hidden opponent made moves that no current program would
>seriously consider, you would be inclined to believe that it was a human player,
>correct?
Agreed.
>If you cannot distinguish between a human or a computer, then any computer that
>could make such a display would be considered to possess intelligence.
>
>I hope I did not bore you with this wholly unoriginal explanation.
Not at all.
What you appear to be saying is that, in the "special" situation of being in a
losing position, one should value a move which creates confusion in the opponent
more highly than the move which scores the highest in terms of leading to the
"least worst" position possible - the normal criteria for evaluation.
You are further saying that the ability to do this is a sign of intelligence.
Given that there is no widely accepted concrete definition of intelligence, I am
inclined to agree with you - especially if the system takes the nature of the
opponent into account.
Sometimes, computers swindle a draw when they are beaten by using their own form
of intelligence - finding moves that the opponent had overlooked. Witness games
3,4, and 5 in DB v GK. This is a frequent experience for me when I play
computers as well.
>Am I a computer generated post, or a human generated one, Graham?
Clearly human. Which some members of the species might find unfortunate... :-)
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.