Re: [computer-go] An [open] question on game tree search theory

[Daniel Hallmark <dgh@xxxxxxxxxx>]

On 19 January 2005, Daniel Hallmark wrote:

> May I ask a question?
> 
> You have defined G(t) as consisting of only 3 discrete values:
> Black wins, draw, or White wins.

snip

> However, and this is most likely my ignorance of Game Theory showing,
> can it be conclusively shown that your definition for G(t) is correct
> for the game of Go?


Thanks to Antti Huima and Tom Cooper for your replies to my message.
As I suspected, my confusion was entirely due to my ignorance of
game theory.  I started doing some background reading after posting
yesterday, and your answers have helped consolidate that information
for me.

I now see that due to back-propagation of leaf values (end-game
results) all previous nodes must have a best- or worst-case value
depending on whose turn it is to play.

Due to its deterministic nature, there is/are forced lines of play
yielding a best result for black from the starting empty board
position.  We simply do not know whether that result is a win,
loss, or draw, and do not know which line(s) of play lead to that
result due to our inability to search the full game tree.

Please pardon the interruption, and thank you for taking the time
to respond to my rudimentary question.

Daniel Hallmark
_______________________________________________
computer-go mailing list
computer-go@xxxxxxxxxxxxxxxxx
http://www.computer-go.org/mailman/listinfo/computer-go/