neural networks for pawn-structures

[romain slootmaekers]

simple short question: who has used neural network stuff to evaluate basic positions (fe pawn endings) and to what result?

[Andreas Wicker]

romain slootmaekers (romain.sl...@student.kuleuven.ac.be) wrote:

: simple short question: who has used neural network stuff to evaluate basic positions (fe pawn endings) and to what result?

AFAIK : H. Braun from the University of Karlsruhe has tried to use NNs for
pawn endings. I don't know how good the results were.

Bye, Andreas
--
Andreas Wicker, Mail:wic...@mvmap66.ciw.uni-karlsruhe.de
So far as the laws of mathematics refer to reality, they are not
certain, and so far as they are certain, they do not refer to
reality. (A. Einstein)