Re: computer-go: A problem with understanding lookahead

[Don Dailey <drd@xxxxxxxxxxxxxxxxxxxxx>]

> No offense, but this is stupid.  It is completely clear that
> n-ply search with random evaluation always plays a uniform
> random move.
> 
> On Thu, 18 Jan 2001, Vincent Diepeveen wrote:


Actually,  even though it defies  common sense, it's true.  A possible
explanation for this  is that the  computer plays to maximize it's own
options  and to minimize the  opponents options, because this makes it
easier to find a  pathway to a  node that scores  highly (gets a  high
random number.)  It chess, it is known that  MOBILITY is a pretty good
thing, if you have a lot more moves than  your opponent you very often
have an advantage.

The random algorithm plays very badly, but the point  is that is still
plays MUCH BETTER than totally random moves.

Don