Re: Fuseki and Joseki Database with Neural Network

[Juergen Kahnert <Kahnert@xxxxxxxxxxxxxxxx>]

DPM> [...] Then by trying all moves, evaluating each,
DPM> and taking the best few to one more level, we can start to find the
DPM> "best" sequence of moves.  However, such an evaluation function
DPM> (score) is very difficult to find, but requires no "steady course"
DPM> property.

You are right. By trying all(!) moves, we need no continuous function,
because we have all possible positions.
But by using GA, we approximate to the maxima or a local maxima. GA find
this quickly while vary the step width - in the beginning big steps and
near a maxima reducing it.
We have to use a step width of one and have no longer benefits from GA.
Thats my understanding of GA or am i wrong?

Juergen