Re: fundamental problems for reinforcement

[Heikki Levanto <heikki@xxxxxx>]

: Ives Steglich schrieb:

: > ...
: > so have thought about some implementation strategies
: > in relation with reinforcement learning end detected some basic
: > problems:
: > 
: > a) i need a way to detect the end of a game
: > b) to rate it in that way that i can say which play has won
: > 
: > this to statements are most importend to get known couse without 
: > this a implementation of a client could get very difficult,

I would think that for reinforcement (or other self-play) learning, it
should be sufficient to say that
  - the game is over when both players pass
  - anything on the board is alive
  - only fully surrounded territory counts
    (that is, any point that can see both colors of stones
    is considered nobody's territory)

This is not sufficient to provide interesting play with humans, and
certainly lacks style and elegance, but for some sort of training it should
be sufficient. Once you have the program trained (how ever you do that), it
might be possible to hack some dirty code to make it pass earlier or to
resign...

Does anybody have a better simple algorithm for the game-end problem??

--
Heikki Levanto  LSD - Levanto Software Development   <heikki@xxxxxxxxxxxxxxxxx>