[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: computer-go: Evaluating positions



The tree branching factor  in 9x9 is  above  70 for the first  several
moves of the game.  For chess it is below 40.  Later in the game, both
boards  get simpler.  I  believe that 7x7 is  a lot closer to chess in
complexity, not 9x9.

Keep us informed on your  progress.  I am  also a  beginner at Go  and
have  written  a  Go program    which I rarely   get  to  work on, but
occasionally play around with.

Don





   Date: Mon, 18 Jun 2001 13:47:02 +0200
   From: Rafael Garcia Leiva <rafael@xxxxxxxxxxxxxxxxx>
   Content-Type: text/plain; charset=us-ascii
   Content-Length: 906


   Dear group:

   I am also working in my own evaluation function, and I was wondering if
   somebody knows where I can find professional games or positions
   evaluated by experts. I would like to compare the result of my
   evaluation function, with what an expert has to say about the position.

   Also I have a question. If the tree-branching-factor of a 9x9 go board
   is simmilar to tree-branching-factor of chess, why the go programs do
   not play better in a 9x9 board? Is it a problem of the evaluation
   function? Why?

   By the way, I am a begginer go player, and I have just started to code
   my own go program.

   Thanks in advance.

   -- 
    ------------------------------- ------------------------------------
   |     Rafael A. Garcia Leiva    |      MS Computational Science      |
   |       rafael@xxxxxxxxxxxxxxxxx      |     Universiteit van Amsterdam     |
    ------------------------------- ------------------------------------