Re: brute force and knowledge

[Robert Jasiek <jasiek@xxxxxxxxxxxxxxx>]

> why are human go experts so unable to express their knowledge ?

Sorry, if I forget about humility this time: I would not have
started a CG project without having this ability. However,
judging from books and talks, most professionals do not seem to
have good analytical abilities. E.g. no one of them dares to define
influence, AFAIK.

> How then could it work ?

You are using some physical model about density, distances, history?
Combined with self-learning mechanisms this might be a promising
approach in the long run, who knows... At least I think that is what
GA and NN programmers should do instead of applying methods to human
go knowledge representation. Why? Physics is much simpler and can be
easily tested with pro games.

-- 
robert jasiek
http://www.snafu.de/~jasiek/