Computer Scrabble significantly exceeds humans. A basic monte carlo search
and an endgame solver is very effective. There is probably still much
strength to be gained (very little opponent modeling is done), but it's
already so strong I don't think it's getting much attention.
Looks like the
I'd be curious on the size of the captures during the game. Imagine
capturing a 1 stone dragon!
- Original Message -
From: "Chris Fant" <[EMAIL PROTECTED]>
To: "computer-go"
Sent: Tuesday, February 20, 2007 10:32 PM
Subject: Re: [computer-go] Big board
this is very interestin
Lukasz, any chance you can access the assembly to see how the compiler
optimized it differently? That is a strange result. All that I could think
of is, if you have several places where the function is called, and the
compiler used to be inlining it, it may be faster (for cache reasons) to
ma
It seems that there are at least three cases:
1: Choosing a random move from a uniform distribution
2: Choosing a random move from a nonuniform distribution (patterns etc.)
3: Choosing a move taking into account what has been chosen before
The concensus seems to be that numbers 1 and 2 are MC a
Are you using the union-find algorithm with path compression for joining
strings? It's very simple and very fast.
http://en.wikipedia.org/wiki/Disjoint-set_data_structure#Disjoint-set_forests
The other main thing to consider is reducing branch mispredictions, which
(I'm guessing) could very we
It's possible to enumerate legal positions exactly:
http://homepages.cwi.nl/~tromp/go/legal.html
Any legal position can be obtained if players are allowed to pass... just
put down the black stones while white passes, then put down the white stones
while black passes.
--Luke Gust
I think you're right, there's the 8 board symmetries and white/black
symmetry, but the latter symmetry is broken since you need to know which
player's turn it is.
Btw, the number of positions that must be considered is also multiplied by
ko considerations.
--Luke Gust
