Re: Fuseki and Joseki Database with Neural Network

["Dean P. McCullough" <dpmccul@xxxxxxxxxxxxxxx>]

> Ok, i exaggerates with the used time span for the evaluation function.
> ;-)
> 
> But i think, the time will not be the main problem. Rather to find a
> really good evaluation function which will work for genetic algorithm.
> I mean, it is impossible to find one.
> Why?
> For GA we need a function with a steady course (i don't know if it is
> the correct mathematical expression for a function without any 'jumps').
> We need a function with minima and maxima. We are able to avoid to end
> in a local maxima by vary the step width, but this requires such a
> steady function.
> And thats the problem. I think we can't find such a function because in
> "go" it's possible a stone placed on a field maybe bad but the field
> next to it could be very good. It is not a little bit better or
> something like that, it could be like day and night. That produce a
> 'jump' in the function.
> Now a generation near the best place would die because it's current
> place is very bad. The solution is to make steps with one field, but
> this means we have to try all the possible fields and this means the
> advantage of GA is lost.
> 

I believe you have the wrong point of view.  Though it would be nice
to have a function which we could maximize to find the "best"
position, what is normally desired is a function which will "reflect'
how good this position is.  Then by trying all moves, evaluating each,
and taking the best few to one more level, we can start to find the
"best" sequence of moves.  However, such an evaluation function
(score) is very difficult to find, but requires no "steady course"
property.   

-- 

Dean P McCullough                            dpmccul@xxxxxxxxxxxxxxxxx