You mean adjusting komi after each simulation or after each move?
(I think the former?)
Yes, each simulation, like this,
play w e5
genmove b
AdjustingKomi(latest_komi);
move = getBestMoveUCT();
latest_komi = CalculateAdjustingKomi();
=c3
play w d4
genmove b
AdjustingKomi(latest_komi);
mov
On Thu, Feb 11, 2010 at 11:24:25PM +0900, Hiroshi Yamashita wrote:
> I also get a bit good result to change komi inside.
> My code is like this.
Interesting, thank you! I already started experimenting with something
quite similar this week too (no commit pushed out in pachi codebase yet),
I call t
Jean-loup Gailly wrote:
m is my definition of the value of the first move.
Willemien wrote:
in 7x7 go under perfect play and without komi black wins by 9 points.
if Black passes on the first move then under the same perfect play
white wins with 9 points (again without komi)
Does that mean th
> All this discussion made me think - has anyone tried to adjust the komi
> between the simulations. Run (say) 10% of the simulations you expect to run
> for a move, and see how many of the moves ended in a win. If there are many
> moves, adjust the komi to make winning harder. If there no moves at
Lets make it al a bit more practical
in 7x7 go under perfect play and without komi black wins by 9 points.
if Black passes on the first move then under the same perfect play
white wins with 9 points (again without komi)
Does that mean that the value of the first move is 18 points?
2010/2/11 J
> Yours is not a proof because what follows is not just a single move
> of value x but a game tree of moves of various sizes,
So let me try to be more precise.
Assume a 19x19 no-handicap game played by perfect players with
New Zealand rules with komi k. k is chosen as the largest value N+0.5
(wit
go.org [mailto:computer-go-
> boun...@computer-go.org] On Behalf Of Petr Baudis
> Sent: Thursday, February 11, 2010 4:52 AM
> To: computer-go
> Subject: Re: [computer-go] Dynamic Komi's basics
>
> Hi!
>
> I forgot two important points:
>
> On Thu, Feb 11, 20
Jean-loup Gailly wrote:
I would write the proof as follows.
Assume x is the value of one move
Yours is not a proof because what follows is not just a single move of
value x but a game tree of moves of various sizes, which need not even
decrease constantly. Many years ago, Barry Phease was a
2010/2/11 Jean-loup Gailly
> A move early in the game is worth about 14 points, not 7.5.
>
While this may be true for professional-level play, the value of the first
move for balancing Monte-Carlo playouts towards a 50% win rate should be
expected to be lower.
Erik
_
Le 11/02/2010 à 16:27, Le Hir Matthieu a écrit :
> In other words, pachi's dynamic komi should be twice as high as it is ?
> Currently, on first move - at 9 handi - pachi has an advantage of about 66
> points thanks to his komi. It should be about 130? How would that affect his
> play ? More second
u
>
>
>
> *---Message original---*
>
> *De :* Jean-loup Gailly
> *Date :* 11/02/2010 16:00:33
> *A :* computer-go
> *Sujet :* Re: [computer-go] Dynamic Komi's basics
>
> > Pachi uses 7.5 points per handicap stone
>
> Pachi is wrong. See
> Pachi uses 7.5 points per handicap stone
Pachi is wrong. See the first paragraph of "There is a relationship"
in http://senseis.xmp.net/?Komi/valueOfFirstMove
The rest of the page is quite confusing but but the value of one move
at the beginning of a game is definitely twice the komi.
A move ear
of handicap games.
Hiroshi Yamashita
- Original Message -
From: "Petr Baudis"
To: "computer-go"
Sent: Thursday, February 11, 2010 9:51 PM
Subject: Re: [computer-go] Dynamic Komi's basics
Hi!
I forgot two important points:
On Thu, Feb 11, 2010 at 01:06:
On Thu, Feb 11, 2010 at 01:06:34PM +0100, Petr Baudis wrote:
> extra_komi = 7.5 * handicap_stones_count
> Then it is linearly decreased until it hits 0 at move 200.
All this discussion made me think - has anyone tried to adjust the komi
between the simulations. Run (say) 10% of the simulations you
Hi!
I forgot two important points:
On Thu, Feb 11, 2010 at 01:06:34PM +0100, Petr Baudis wrote:
> > On the other hand, 9 handicaps are supposedly giving an advantage of 90 to
> > 120 points, so my natural thought would be that the bot would give itself at
> > least a negative komi of that man
The basic i think of Dynamic komi in my opinion is the following.
The bot must be able to distinguish good from bad moves (this doesn't
need much explanation i guess)
and a bot '' by implementation" thinks that the opponent is as good as
he is. (in most algorithms he is playing against himself)
T
Hi!
On Thu, Feb 11, 2010 at 10:32:49AM +0100, Le Hir Matthieu wrote:
> First, I'm wondering how komi is determined when a dynamic system is used :
>
> * According to this page : http://senseis.xmp.net/?Komi%2FvalueOfFirstMove
> the value of komi at first move is half the points the black move is
Hi
Le 11/02/2010 à 10:32, Le Hir Matthieu a écrit :
> Hi,
>
>
> From what I think I understood, dynamic komi is supposed to try to keep the
> game more even.
The dynamic komi is a bias in the evaluation in order to "inform" the bot
that the game is balanced, and prevent it beeing "blinded" by
Le Hir Matthieu wrote:
First, I'm wondering how komi is determined when a dynamic system is
> used
Either use komi bidding or determine a new (or the same) komi value for
every next game. For the latter, the basic idea is to observe for which
komi the winning rates were above / below 50%. You
try: (handicap - 0.5) x 14
Erik
2010/2/11 Le Hir Matthieu
>Hi,
>
>
> I have a few questions concerning dynamic komi, I am not a programmer
> though and will try my best to be understandable.
>
> First, I'm wondering how komi is determined when a dynamic system is used :
>
> * According to
20 matches
Mail list logo
