http://www.theguardian.com/science/brai ... test-spoof
What about a Turing test for chess?
You play blitz against a program for 2 hours, and you are convinced you are playing against a human with 1800 rating.
When is that possible?
Which program is the best to simulate a human blitz player today?
How long will it take to determine it is a computer?
More info about human play in this and related threads:
http://talkchess.com/forum/viewtopic.ph ... ew=threads
A Turing blitz chess test
Moderators: hgm, Rebel, chrisw
-
JBNielsen
- Posts: 267
- Joined: Thu Jul 07, 2011 10:31 pm
- Location: Denmark
-
aturri
- Posts: 85
- Joined: Wed Dec 30, 2009 11:35 pm
Re: A Turing blitz chess test
The claim seems not valid:JBNielsen wrote:http://www.theguardian.com/science/brai ... test-spoof
https://www.techdirt.com/articles/20140 ... tter.shtml
I think it is much easier to detect a computer playing at full/near full strength (>2700 human ELO) that at lower level. I don't think there is unique play for people playing at 1600-2000 ELO: some are good at tactics, other sucks at the endgame, some only know openings, others are not interested at all on theory, some just play for fun and sacrifying, many of us just don't have any strategic plan during the whole game, so the moves seems disconnected between them, etc.What about a Turing test for chess?
You play blitz against a program for 2 hours, and you are convinced you are playing against a human with 1800 rating.
When is that possible?
Which program is the best to simulate a human blitz player today?
How long will it take to determine it is a computer?
More info about human play in this and related threads:
http://talkchess.com/forum/viewtopic.ph ... ew=threads
Due the fact that present IA researches are not interested on passing the real Turing test, I doubt they would be interested on trying the chess approach.