Re: [computer-go] Pattern matching - example play

[Vincent Diepeveen <diep@xxxxxxxxx>]

At 10:34 29-11-2004 -0800, David Weiss wrote:
>One can look at "Go World" or similar publications
>that analyze professional games.  They are filled
>with diagrams of alternate moves that were considered
>and rejected, along with the reasons for that.

Play some 65 years old player who some years ago played for the world title
chess.

They just see each time it is their move a single move. As they do not see
a single other move and just consider this move they play it. That move
happens each time to be the best move, so they win the game.

Then after the game they write for some magazine an analysis of their game
and they find incredible alternative lines which did not work for them.

I remember that i played Victor Korchnoi a few years ago. Start of the 70s
when the cold war was at its peek, Victor Korchnoi (switzerland) played an
important match against Anatoly Karpov (SSSR) for the world title.

In those days such matches were politically very loaded.

Of course i played him somewhere in the 90s when the tension was no longer
there and Korchnoi with 2690 was no longer number 2 of the world, but
roughly number 20 of the world (i have to see a second player doing that at
an age that normally people retire).

During the analysis with me, with 30 reporters standing around the board.

Then Victor forced my side to play a few horrible moves which i would not
consider over my dead body.

Trivially the result of all that was that my position was dead lost and
then Victor could show for 30 minutes to the reporters line after line
which of course all worked out real well for the white side (i don't need
to mention that Victor played white). Of course great variations which then
were published in some magazines.

That Victor really had seen a lot... ...in the analysis

Reality is this. In netherlands the name 'De Groot' is like the name 'Ye'
in China, or Muralidharan in India. In 1946 a researcher called De Groot
has very clearly proven that professional players do not search more lines
than strong amateur players do. So they do not win games because they see
MORE. They win games because they see the same but consider BETTER moves.

>This has been pointed out before on this mailing list.
>The moves that are actually played are the results
>of many moves being rejected that are not seen in
>professional or strong amateur games.  
>
>A program that attempts to predict pro moves will
>have a hard time when the opponent makes moves that
>would be terrible against a strong player.
>
>
>> A simple answer: trick moves. There are lots of
>> tricky sequences that
>> professionals never play, because they get a result
>> that is a little bit
>> less optimal than what they can get. yet such moves
>> often offer many
>> ways for a less experienced opponent to screw up.
>> Such moves do not
>> occur in pro games, but often amateurs use them,
>> especially against
>> weaker players, and with great success.
>> 
>> Same kind of thing happens when people try to
>> memorize joseki - they can
>> handle players who stick to josekis, but are totally
>> lost when a player
>> deviates from the book line, because they do not
>> understand the meaning
>> behind every move in those josekis.
>> 
>
>
>
>	
>		
>__________________________________ 
>Do you Yahoo!? 
>Yahoo! Mail - You care about security. So do we. 
>http://promotions.yahoo.com/new_mail
>_______________________________________________
>computer-go mailing list
>computer-go@xxxxxxxxxxxxxxxxx
>http://www.computer-go.org/mailman/listinfo/computer-go/
>
>
_______________________________________________
computer-go mailing list
computer-go@xxxxxxxxxxxxxxxxx
http://www.computer-go.org/mailman/listinfo/computer-go/