> I am running some public available bots on KGS now and than ( MoGo,
> GnuGo, Fuego, ...) under Windows (precompiled builds).
>
> Where may i get information about the format of the opening databases
> of these bots ? Is there a common format for opening databases so far
> ?
I don't think there
Hi, im new to this list and i am no develloper.
I am running some public available bots on KGS now and than ( MoGo, GnuGo,
Fuego, ...) under Windows (precompiled builds).
Where may i get information about the format of the opening databases of these
bots ?
Is there a common format for opening d
On Thu, Jun 18, 2009 at 10:20 AM, Brian Sheppard wrote:
> >> In my old Goliath program I used positions as hashcodes to lookup a
> >> move/value combination because it was much easier to program than
> >> trees :)
>
> >I think this is the better solution and it's probably easier too.
>
> You prob
>> In my old Goliath program I used positions as hashcodes to lookup a
>> move/value combination because it was much easier to program than
>> trees :)
>I think this is the better solution and it's probably easier too.
You probably need both an automatically generated system (self-learning)
and a
On Wed, Jun 17, 2009 at 8:37 PM, Mark Boon wrote:
> On Wed, Jun 17, 2009 at 9:17 AM, David Fotland
> wrote:
> > Many Faces uses position value collection. Positions are hashed and
> looked
> > up in the position table (with a hash invariant to rotations and
> > reflections). Each node has infor
On Wed, Jun 17, 2009 at 9:17 AM, David Fotland wrote:
> Many Faces uses position value collection. Positions are hashed and looked
> up in the position table (with a hash invariant to rotations and
> reflections). Each node has information about the position (wins, losses,
> strongest players at
I
switched to the position/value collection in version 12.
David
-Original Message-
From: computer-go-boun...@computer-go.org
[mailto:computer-go-boun...@computer-go.org] On Behalf Of Willemien
Sent: Wednesday, June 17, 2009 4:45 AM
To: computer-go@computer-go.org
Subject: [computer-go] open
> 1 a position -> move collection (if the position is such make that move)
> 2 a position value collection (this position is good, that position is
> bad for the colur who is on the move)
> 3 a professional game collection (let the program just play like the
> pro's)
> 4 a game tree (game with lo
(I just forgot to mention that we take care of rotations / symetries)
I was puzzeling what is the best way to organise an opening book?
>>
>
> Our solution was the following. We have two possible structures:
> 1) a big set of SGF games;
> 2) an opening book in which one line contains p,w,l where
>
I'm assuming small boards. For big boards it's not practical to have a
database style book of course.
You definitely must take into considerations board orientations. I suggest
using a database where the position is hashed. You can do your own or you
can use a lite sql database such as sqlite
Are you asking about in-memory storage, or on disk?
On disk, you probably want to make sure that it's as easy as possible to
maintain - up to your taste. Some people like a single SGF file with
variations. You can do mirrorring and transpositions when you load the book.
Christian
Willemien
>
> I was puzzeling what is the best way to organise an opening book?
>
Our solution was the following. We have two possible structures:
1) a big set of SGF games;
2) an opening book in which one line contains p,w,l where
* p is a situation
* w is the number of wins
* l is the number o
2009/6/17 Willemien :
> I was puzzeling what is the best way to organise an opening book?
It depends of how you program will play.
> i am writing a program to be very strong on 7x7 go program (as
> prelimary for writing a very strong 9x9 program ) and am wondering
> what is the best structure for
I was puzzeling what is the best way to organise an opening book?
i am writing a program to be very strong on 7x7 go program (as
prelimary for writing a very strong 9x9 program ) and am wondering
what is the best structure for an opening book.
Where i am at the moment is:
There are about 4 funda
Those are some interesting links. Does anyone know the prediction rates
of kombilo?
On Wed, 2007-10-17 at 22:13 +0200, Rémi Coulom wrote:
> Erik S. Steinmetz wrote:
> > I would like to thank everyone who responded on this thread. The
> > pointers have been very helpful.
> >
> > I would also lik
Erik S. Steinmetz wrote:
I would like to thank everyone who responded on this thread. The
pointers have been very helpful.
I would also like to see that linked document, as the text describing
the pattern value system looks interesting, and a longer description
of it would be nice! If anyone
Frank responded immediately, and posted the document. It is now
accessible via the link in the help files.
http://moyogo.com/MoyoGoPatternSystem.pdf
The document contains a bit, but not much, more than is available in
the description in the help files.
On Oct 17, 2007, at 1:04 PM, Erik S.
I would like to thank everyone who responded on this thread. The
pointers have been very helpful.
I would also like to see that linked document, as the text describing
the pattern value system looks interesting, and a longer description
of it would be nice! If anyone can find a copy, please
Le lundi 15 octobre 2007 20:53, Erik S. Steinmetz a écrit :
> Greetings all,
>
> I have been looking through the literature (many thanks to Markus's
> wonderful online bibliography) on existing strategies in the opening
> game, and have not found too many articles on the specifics outside
>
I'm glad you posted about that. I was thinking of it but couldn't remember
the name of it.
It looks like there may have been some data loss in the past (discussed in
blogs). I see several links that are broken. I found the text at
http://www.moyogo.com/files/%7BAF6C0FD3-B2EF-4CBE-85CB-7359B82A4
On 10/15/07, Erik S. Steinmetz <[EMAIL PROTECTED]> wrote:
> Greetings all,
>
> I have been looking through the literature (many thanks to Markus's
> wonderful online bibliography) on existing strategies in the opening
> game, and have not found too many articles on the specifics outside
> of a few
in figure 6 of this paper, black has developed (but hasn't
yet used) the mother of all walls. it's pretty funny.
s.
- Original Message
From: terry mcintyre <[EMAIL PROTECTED]>
To: computer-go
Sent: Monday, October 15, 2007 4:17:47 PM
Subject: Re: [computer-go] Opening
Webster
>
> - Original Message
> From: Erik S. Steinmetz <[EMAIL PROTECTED]>
> To: computer-go
> Sent: Monday, October 15, 2007 11:53:25 AM
> Subject: [computer-go] Opening game strategies
>
> Greetings all,
>
> I have been looking through the literature (ma
ean to govern. They promise to be kind
masters; but they mean to be masters. -- Daniel Webster
- Original Message
From: terry mcintyre <[EMAIL PROTECTED]>
To: computer-go
Sent: Monday, October 15, 2007 12:45:14 PM
Subject: Re: [computer-go] Opening game strategies
http://stat.cs.
t they mean to govern. They promise to be kind
masters; but they mean to be masters. -- Daniel Webster
- Original Message
From: Erik S. Steinmetz <[EMAIL PROTECTED]>
To: computer-go
Sent: Monday, October 15, 2007 11:53:25 AM
Subject: [computer-go] Opening game strategies
Greet
Thanks! That definitely looks like a topical paper that I managed to
overlook.
On Oct 15, 2007, at 2:02 PM, Jason House wrote:
Have you checked out what Remi Coulom did?
http://remi.coulom.free.fr/Amsterdam2007/
It doesn't claim to be the best in the opening, but does claim to
do a decent
Have you checked out what Remi Coulom did?
http://remi.coulom.free.fr/Amsterdam2007/
It doesn't claim to be the best in the opening, but does claim to do a
decent job in the opening (and a better job in the middle of the game)
On 10/15/07, Erik S. Steinmetz <[EMAIL PROTECTED]> wrote:
>
> Greeting
Greetings all,
I have been looking through the literature (many thanks to Markus's
wonderful online bibliography) on existing strategies in the opening
game, and have not found too many articles on the specifics outside
of a few papers on neural net learning applied to the opening. There
My program only stores 1 canonical position in symmetrical instances.
But that is only for the opening book. Where there is a choice of
symmetrically equivalent moves, it makes one at random.
For instance, there might only be one canonical response to e5 in the
opening position but this could
But I would still like to know how many MC evaluations it would take
until all corners look at least somewhat similar... I bet that is many
more than we see used currently - if it ever gets there.
The approach I take in Valkyria is radically different. I simply prune all
symmetric moves. For e
on 9x9 it's easier to see it converge. 19x19 is a beast,
which is why i think that scanning a small slice of the board
for the first two moves might not be such a bad idea.
s.
Fussy? Opinionated? Impos
From: Heikki Levanto <[EMAIL PROTECTED]>
> I still think it a bit strange that on an empty board, a program can
> prefer a 3-3 point in one corner, and in another corner find it quite
> unplayable.
It makes sense of the space evaluated by the random playouts differed. But my
thinking
is, what
> For the first few moves perhaps, but after that, symmetric positions
> must be awfully rare, and worrying about them probably costs more than
> it can ever give.
it's pretty quick to check for, but i agree, it's likely not terribly useful.
however, i do think that it'd be very, very helpful for
On Sat, Jun 16, 2007 at 02:29:51PM -0400, [EMAIL PROTECTED] wrote:
> My program tends to amplify noise fluctuations. If it's presented with
> several equally good moves, it will pick winners and losers early on.
Actually, the more I think about this, the more sense it makes to me. If
a program fin
only for the first move or three, really.
s.
- Original Message
From: terry mcintyre <[EMAIL PROTECTED]>
To: computer-go
Sent: Monday, June 18, 2007 1:09:31 PM
Subject: Re: [computer-go] Opening
Is it possible to recognize and exploit symmetry to improve the quality of the
On Mon, Jun 18, 2007 at 10:09:31AM -0700, terry mcintyre wrote:
>
> Is it possible to recognize and exploit symmetry to improve the
> quality of the move estimation process with minimal expenditure of
> effort?
For the first few moves perhaps, but after that, symmetric positions
must be awfully r
asters. -- Daniel Webster
- Original Message
From: Jacques Basaldúa <[EMAIL PROTECTED]>
To: computer-go@computer-go.org
Sent: Sunday, June 17, 2007 4:16:11 AM
Subject: [computer-go] Opening
Heikki Levanto wrote:
> I am sure there is a mathematically sound way to measure
> how symmet
Heikki Levanto wrote:
> I am sure there is a mathematically sound way to measure
> how symmetric the evaluation is, but my math is a bit rusty,
> so I am asking if someone can come up with a good way. After
> that, I'm asking if various programmers would be willing to
> run this test, and publish
t;[EMAIL PROTECTED]>
To: computer-go
Sent: Sat, 16 Jun 2007 6:48 am
Subject: [computer-go] Opening
It seems generally accepted that MC or UCT programs are weakest in the
pening. My own experience matches this too. Some times I get the idea
hat my program doesn't know at all what it is doing
the point equivalence classes are easy enough to define. points are
either in a size-1 class (the center point, can be ignored), a size-4
class (axes, both vert/horiz. and diagonal), or a size-8 class (all other
points on the board).
for each tuple of points in the equivalence class, normalize wh
There are two distinct concepts here. There is precision and there is
accuracy. Your test will test precision (how exact the estimate is), but it
fails to test accuracy (a measure of how close the mean is to the actual
value). But, I do think that knowing the precision is useful. If the
pre
It seems generally accepted that MC or UCT programs are weakest in the
opening. My own experience matches this too. Some times I get the idea
that my program doesn't know at all what it is doing the first few
moves. I propose a simple test to see if that is the case. Before doing
it, I'd like to he
42 matches
Mail list logo
