[computer-go] refutations useful

[terry mcintyre]

The trouble with emulating good moves in pro games is that we don't know why those moves are good.

Lately, I've been studying joseki, and I find that it's hard to really know a joseki until you know why non-joseki moves are bad - and why moves which are locally joseki may be bad in relation to other stones on the board.
 
Maybe we need to add some examples of refutations to our training games.  For instance, I am looking at "Tricks in Joseki", by Yang Yilun -- he shows a variety of non-joseki moves, and their appropriate refutations. One may see such moves often, sometimes played by players hoping to test your understanding of joseki, sometimes by players who are themselves groping to understand the position.

I've been observing games by players who are much stronger than I am. Interestingly, I can predict at least 80% of their moves. Such predictive abilities in a computer program would be considered quite respectable. Nonetheless, these players can beat me with a five-stone handicap more often than not. My predictive abilities are shallow; in certain crucial areas, my blunders cost me five points here, ten points there. A handful of such blunders will negate the advantage of a five stone handicap.




Terry McIntyre <terrymcintyre at yahoo.com>
They mean to govern well; but they mean to govern. They promise to be kind masters; but they mean to be masters. -- Daniel Webster




       
____________________________________________________________________________________Ready for the edge of your seat? 
Check out tonight's top picks on Yahoo! TV. 
http://tv.yahoo.com/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://computer-go.org/pipermail/computer-go/attachments/20070705/20b0225a/attachment.htm