On Thu, Feb 18, 2010 at 7:28 AM, Ingo Althöfer 3-hirn-ver...@gmx.dewrote:
Hello Don,
several very good points by you!
Does anyone have data based on several thousands games
that attempts to measure the effect of dynamic komi?
I would like to see results that are statistically
Baudis pa...@ucw.cz
On Wed, Feb 17, 2010 at 05:29:36PM -0500, Don Dailey wrote:
Does anyone have data based on several thousands games that attempts to
measure the effect of dynamic komi?I would like to see results that
are
statistically meaningful. We need to see a few thousand games
2010/2/18 dhillism...@netscape.net
Ingo,
I'm not a proper statistician, but I believe there's a crucial second step
that's missing in your analysis of significance. Even if this were the only
computer-go test that you personally had ever conducted, we would
nevertheless need to take into
Does anyone have data based on several thousands games that attempts to
measure the effect of dynamic komi?I would like to see results that are
statistically meaningful. We need to see a few thousand games played
against a fixed opponent WITH dynamic komi, and then the same program
without
That's pretty impressive for the go language if this is an apples to apples
comparison. Is it pretty much?
On Wed, Dec 16, 2009 at 9:50 PM, Brian Slesinsky br...@slesinsky.orgwrote:
Oops, you're right. Here it is with -server:
Plug-and-Go refbot:17857
CRef bot (-O3)
That same statement baffles me. AMAF gives a huge boost with light
playouts for me. As the number of playouts increase, AMAF gives less and
less. After a few thousand playouts it's almost nothing but if it's worse
than not doing AMAF it is difficult to measure. Of course MCTS never does
That's awesome!
Do you have performance numbers on the same hardware for the C refbot?
- Don
On Sat, Dec 12, 2009 at 7:39 PM, Brian Slesinsky br...@slesinsky.orgwrote:
Thought I'd announce that I've ported the Java refbot to the Go
language (with some modifications).
I'm getting about
The empty value is not needed. In some games it's easier to have it
because it can be a simplification - everything handled uniformly for
instance and it can avoid a conditional branch but I don't think that is an
issue with Go.
- Don
On Tue, Dec 8, 2009 at 5:49 PM, Petr Baudis pa...@ucw.cz
A few months ago there was a post in the computer chess forums about
optimizing combinations of features. It was called orthogonal
multi-testing.
Did I mention that on this forum already? If not, here is a brief on how
it works:
Suppose you have 1 feature you want to test - you might
I know there are heuristics for trying to understand the interactions and
without looking too hard I assume this package is just a more comprehensive
version of this.
On Wed, Nov 25, 2009 at 9:11 AM, steve uurtamo uurt...@gmail.com wrote:
the way to do all of this exactly is with experimental
On Wed, Nov 25, 2009 at 10:44 AM, Heikki Levanto hei...@lsd.dk wrote:
On Wed, Nov 25, 2009 at 09:01:22AM -0500, Don Dailey wrote:
You could of course just play games where you choose each player
randomly.
If you have 256 feature you have a ridiculous number of combinations,
more
than you
Berlekamp came to MIT and gave a talk for us, and after that we talked
about Go and Chess and other things and took him out to eat.
I can vouch for the fact that he is a truly humble and modest person and is
a real joy to talk to. It was all thoroughly enjoyable.
- Don
On Wed, Nov 25, 2009
On Wed, Nov 25, 2009 at 2:00 PM, Matthew Woodcraft
matt...@woodcraft.me.ukwrote:
steve uurtamo wrote:
the way to do all of this exactly is with experimental design.
to design experiments correctly that handle inter-term interactions of
moderate degree, this tool is quite useful:
On Mon, Nov 23, 2009 at 1:58 AM, Robert Jasiek jas...@snafu.de wrote:
Don Dailey wrote:
this simplification of the rules
Simplification? It does not even simplify strategy.
I am asserting that a properly modified bot is going to better at this
variant of the game. It's way easier to play
On Mon, Nov 23, 2009 at 9:34 AM, Alain Baeckeroot
alain.baecker...@laposte.net wrote:
Le 23/11/2009 à 15:04, Don Dailey a écrit :
On Mon, Nov 23, 2009 at 1:58 AM, Robert Jasiek jas...@snafu.de wrote:
Don Dailey wrote:
this simplification of the rules
Simplification? It does
I have repeatedly stated that the Hahn system is a simplification, but this
is just a guess on my part and I might have it backwards.I'm not sure
whether that invalidates the idea that computers will play this better or
not.
Here is a thought experiment.Imagine an omniscient player or
On Mon, Nov 23, 2009 at 11:07 AM, Robert Jasiek jas...@snafu.de wrote:
Don Dailey wrote:
I think it's simpler because I am a weak
player and I think more in terms of total points rather than winning
games
Many weak players have told me (and for me when I was a beginner it was the
same
.
- Don
On Mon, Nov 23, 2009 at 12:01 PM, Robert Jasiek jas...@snafu.de wrote:
Don Dailey wrote:
In win game mode [God] will play ANY move randomly that is good enough.
If God is set to play any randomly chosen winning move, yes.
Since it is omnicient there is no point in talking
What I cannot decide is if it is really more
challenging - I just know it's more challenging to do it perfectly.
More challenging for whom? For God, it is equally boring.
More challenging in the sense that more work must be done.
- Don
--
robert jasiek
On Mon, Nov 23, 2009 at 12:01 PM, Robert Jasiek jas...@snafu.de wrote:
Don Dailey wrote:
In win game mode [God] will play ANY move randomly that is good enough.
If God is set to play any randomly chosen winning move, yes.
Since it is omnicient there is no point in talking about risk
On Mon, Nov 23, 2009 at 1:39 PM, Robert Jasiek jas...@snafu.de wrote:
GoGod and GoDevil are objective technical terms referring to the game tree.
They were defined roughly on rec.games.go quite some years ago but I do not
recall the definition details by heart. They have nothing to do with
On Mon, Nov 23, 2009 at 3:23 PM, Nick Wedd n...@maproom.co.uk wrote:
In message 5212e61a0911231136t1e83ce37i9375a033fe3e0...@mail.gmail.com,
Don Dailey dailey@gmail.com writes
On Mon, Nov 23, 2009 at 12:01 PM, Robert Jasiek jas...@snafu.de
wrote:
Don Dailey wrote:
In win game
On Mon, Nov 23, 2009 at 4:51 PM, Robert Jasiek jas...@snafu.de wrote:
Don Dailey wrote:
If all moves lose, how would YOU select?
E.g., I choose some that creates the most ready traps.
Did you get the point that I'm defining 2 separate strategies?One is
to
maximize the points
It's too early for computers to win bets like this - but it will eventually
happen. We just need to wait a few more years which will come and go
before you realize it.
- Don
2009/11/22 terry mcintyre terrymcint...@yahoo.com
Any hardware which can be brought to the playing site would
I'm pretty sure that this simplification of the rules would favor computers.
Of course that would require some program modifications, primarily
counting points on the board instead of wins and losses.
These rules basically takes out some (or at least reduces) elements of the
game that humans
There is no question that computers play better at longer time controls even
though this has been disputed on this group. Is there any issues with
parallelism at short searches?In the old days when I competed in
computer chess with many processors, the program could out-search the
single
2009/10/29 Olivier Teytaud olivier.teyt...@lri.fr
Yes, this group does not have a consensus at all on this. On the one
hand we hear that MCTS has reached a dead end and there is no benefit from
extra CPU power, and on the other hand we have these developers hustling
around for the
2009/10/29 Olivier Teytaud olivier.teyt...@lri.fr
Just curious, who actually claimed that and what was it based on?
I don't know who claimed it first, and who agreed for it,
but I agree with it :-)
But you always seek the most hardware when you play against a human it
seems.
I think
On Thu, Oct 29, 2009 at 12:40 PM, Petr Baudis pa...@ucw.cz wrote:
On Thu, Oct 29, 2009 at 12:00:32PM -0400, Don Dailey wrote:
That is exactly as it should be and is not a barrier. I don't think you
know the difference between a wall and a point that is just far away.
I'd phrase
Yes, I agree with you on most of this. However, I believe that Go is a
very simple domain in some sense and that we romanticize it too much. I am
not saying there is not amazing depth to it, but it's represented very
compactly and it's a game of perfect information with very limited
What is interesting is not the fact that intrasitivity exists, that is not
in doubt. But it quite interesting that this much intransitivity can be
created with non-trivial and strong programs.
I would like to see the data though, specifically the number of games
between each player at each level
2009/10/26 Richard J. Lorentz lore...@csun.edu
How things changes. You would never hear a comment like Remark c) below
concerning the old alpha-beta chess engines.
Yes, this group does not have a consensus at all on this. On the one hand
we hear that MCTS has reached a dead end and there
Peter, did your comment get cut off?
Anyway, I agree with you on this. Humans are not stronger on short time
settings. I believe that SOME humans could be better if they have a
problem staying interested for a longer period of time and the longer time
control upsets their rhythm or
at 6:14 PM, Mark Boon tesujisoftw...@gmail.com wrote:
2009/10/26 Don Dailey dailey@gmail.com:
2009/10/26 Richard J. Lorentz lore...@csun.edu
Yes, this group does not have a consensus at all on this. On the one
hand
we hear that MCTS has reached a dead end
On Wed, Oct 14, 2009 at 10:21 AM, Álvaro Begué alvaro.be...@gmail.comwrote:
We should let go of this idea that artificial neural networks have
anything to do with the brain. ANNs are just a family of parametric
functions (often with too many parameters for their own good) and
associated
One must be very careful about proclaiming wild transitivity issues. I'm
not saying it's not an issue, there is some going on with every program on
CGOS, but with less than 500 games between any two players you are going
to get error margins of +/- 30-50 ELO or something like that.
And CGOS
2009/10/6 Jason House jason.james.ho...@gmail.com
That already exists: kgs-game_over is sent after every game (if you support
it). That it's up to your bot to decide if it should terminate, run a full
garbage collection, pause pondering, etc... I think most people use a
sentinel file.
Are
2009/9/12 jorge jorge...@terra.es
sorry, but as i don´t receive anything, i don´t know is the list is not
active or if i´m doing something wrong ...
It's fairly active, but you might not get a message every single day.
- Don
___
computer-go
I tried both llvm-gcc and CLANG. I did not have any trouble getting them
to work for my 64 bit chess program.
I didn't try too hard, but neither is producing executables as fast as
gcc. llvm-gcc is the slowest about 20% slower than gcc and clang is only a
little slower than gcc.
Since I
A couple of notes. Some of us on this computer-go forum are also Chess
programmers and many of write 64 bit chess programs. I am one of them but
I know there are others.
My 64 bit chess program runs almost 2X faster if I compile it to run on a 64
bit OS - in other words I'm using real 64 bit
I'm glad to see some are actually experimenting with this.
My suggestion is to modify a program such as fuego to follow one of the
algorithms as suggested - then test it with a large sample of games. If
it doesn't work we can experiment until it does or until we are satisfied
that it won't.
One must decide if the goal is to improve the program or to improve it's
playing behavior when it's in a dead won or dead lost positions.
It's my belief that you can probably cannot improve the playing strength
soley with komi manipulation, but at a slight decrease in playing strength
you can
.
- Don
-Magnus
Quoting Don Dailey dailey@gmail.com:
One must decide if the goal is to improve the program or to improve it's
playing behavior when it's in a dead won or dead lost positions.
It's my belief that you can probably cannot improve the playing strength
soley with komi
, but perhaps they do not.
Terry McIntyre terrymcint...@yahoo.com
“We hang the petty thieves and appoint the great ones to public office.” --
Aesop
--
*From:* Don Dailey dailey@gmail.com
*To:* computer-go computer-go@computer-go.org
*Sent:* Wednesday, August 19
PS: Once again I would like to mention my report on Laziness of Monte
Carlo, at http://www.althofer.de/mc-laziness.pdf
In the meantime, a student has found the same phenomenon in UCT search
(instead of basic MC). Also in discrete online optimization (so outside
of combinatorial games) it
On Wed, Aug 19, 2009 at 2:11 PM, Brian Sheppard sheppar...@aol.com wrote:
My conclusion is the same as Gian-Carlo Pascutto's: I am convinced
that the phenomenon of laziness is real, and that it hurts
practical strength.
Unfortunately this is not that point that is in question - I think we
2009/8/15 Jason House jason.james.ho...@gmail.com
On Aug 14, 2009, at 11:02 PM, David Fotland fotl...@smart-games.com
wrote:
Moves often merge two groups.
I count liberties incrementally as I make moves, so no need to search to
count.
How do you detect shared libreties to avoid double
can use the code of one
implementation to debug then next - always checking to see that you get the
same answer.
- Don
2009/8/15 Jason House jason.james.ho...@gmail.com
On Aug 15, 2009, at 8:22 AM, Don Dailey dailey@gmail.com wrote:
2009/8/15 Jason House jason.james.ho...@gmail.com
Have you looked at scala yet?I don't understand Erlang performance but
scala gives you something higher level than Java or C and same performance
as Java, which for most long running applications is pretty close to C
performance.I'm currently taking a look at it - I'm always on the
I don't think JVM performance will be an issue for this.I assumed that
you were willing to sacrifice a small amount of speed for a high level
prototyping language and I think you will only get about 20-30% slowdown
over C - I'm judging this by the performance of the reference bots I did in
On Fri, Aug 14, 2009 at 5:13 PM, Carter Cheng carter_ch...@yahoo.comwrote:
I have been having difficulties selecting a good representation for liberty
sets for strings of stones. I am curious how other people might be doing
this. I suspect that for heavier playouts one would like to know not
--
*From:* Don Dailey dailey@gmail.com
*To:* computer-go computer-go@computer-go.org
*Sent:* Friday, August 14, 2009 2:25:06 PM
*Subject:* Re: [computer-go] Erlang and computer go
I don't think JVM performance will be an issue for this.I assumed that
you were willing
doing it to have some method of
liberty counting + a exhaustive search to determine the last two liberties
for example?
--- On Fri, 8/14/09, Don Dailey dailey@gmail.com wrote:
From: Don Dailey dailey@gmail.com
Subject: Re: [computer-go] representing liberties
To: computer-go
Of *Don Dailey
*Sent:* Friday, August 14, 2009 6:17 PM
*To:* computer-go
*Subject:* Re: [computer-go] representing liberties
I'm not sure I understand your question. But I'll try to explain it a
little better.
Basically, you keep a C structure or the equivalent which tracks each
On Fri, Aug 14, 2009 at 9:51 PM, David Fotland fotl...@smart-games.comwrote:
Old Many Faces keeps linked lists of liberties for each group. They are
sorted, singly linked lists, so merges are fast.
Yes, I can see that merges would be really fast with linked lists. Are
they common enough to
This idea makes much more sense to me than adjusting komi does.At least
it's an attempt at opponent modeling, which is the actual problem that
should be addressed. Whether it will actually work is something that
could be tested.
Another similar idea is not to pass but to play some
On Thu, Aug 13, 2009 at 1:39 AM, Christoph Birk b...@ociw.edu wrote:
On Aug 12, 2009, at 3:43 PM, Don Dailey wrote:
I believe the only thing wrong with the current MCTS strategy is that you
cannot get a statistical meaningful number of samples when almost all games
are won or lost.You
There is one crude way to measure goal compatibility. See if you can make
the same move work with different komi.If I'm on the east coast of the
US traveling to the west coast, I will probably start off on the same road
regardless of whether I'm going to Seattle or San Diego.If the same
.( Or if every move satisfies the long
term goal in case of taking handicap)
Stefan
- Original Message -
*From:* Don Dailey dailey@gmail.com
*To:* tapani.ra...@tkk.fi ; computer-go computer-go@computer-go.org
*Sent:* Thursday, August 13, 2009 4:02 PM
*Subject:* Re: [computer-go
office.” --
Aesop
--
*From:* Don Dailey dailey@gmail.com
*To:* computer-go computer-go@computer-go.org
*Sent:* Thursday, August 13, 2009 9:27:11 AM
*Subject:* Re: [computer-go] Dynamic komi at high handicaps
2009/8/13 Stefan Kaitschick stefan.kaitsch
Yes, known problem :-( I'm still trying to find a method to see if a
point is in an eye. Should not be too difficult in theory but in
practice i have not found a method yet.
Are you talking about 1 point eyes? For this I think most programs use
the same definition, which is quite good
2009/8/12 Ingo Althöfer 3-hirn-ver...@gmx.de
In the last few weeks I have experimented a lot with dynamic
komi in games with high handicap. Especially, I used the
really nice commercial program Many Faces of Go (version 12.013)
with its Monte Carlo level (about 2 kyu on 19x19 board) and
its
Ok, I misunderstood his testing procedure. What he is doing is far more
scientific than what I thought he was doing.
There has got to be something better than this. What we need is a way to
make the playouts more meaningful but not by artificially reducing our
actual objective which is to
The problem with MCTS programs is that they like to consolidate. You set
the komi and thereby give them a goal and they very quickly make moves which
commit to that specific goal. Commiting to less than you need to actually
win will often involve sacrificing chances to win.Sometime it
in a field which would
otherwise look like a dark, desolate, win-less landscape.
Terry McIntyre terrymcint...@yahoo.com
“We hang the petty thieves and appoint the great ones to public office.” --
Aesop
--
*From:* Don Dailey dailey@gmail.com
*To:* computer-go
On Wed, Aug 12, 2009 at 5:36 PM, Mark Boon tesujisoftw...@gmail.com wrote:
I started to write something on this subject a while ago but it got
caught up in other things I had to do.
When humans play a (high) handicap game, they don't estimate a high
winning percentage for the weaker player.
2009/8/12 terry mcintyre terrymcint...@yahoo.com
Most experiments are done on even games; this dynamic algorithm applies
particularly to handicap games.In that context, it is not an ungainly
kludge, but actually reflects the assessment of evenly matched pro players -
they look at the board,
On Wed, Aug 12, 2009 at 5:58 PM, Mark Boon tesujisoftw...@gmail.com wrote:
2009/8/12 Don Dailey dailey@gmail.com:
I disagree about this being what humans do. They do not set a fake komi
and then try to win only by that much.
I didn't say that humans do that. I said they consider
On Wed, Aug 12, 2009 at 6:03 PM, Matthew Woodcraft
matt...@woodcraft.me.ukwrote:
Don Dailey wrote:
The problem with MCTS programs is that they like to consolidate. You
set the komi and thereby give them a goal and they very quickly make
moves which commit to that specific goal.
How did
. Handicap go wasnt given special attention sofar.
Stefan
- Original Message -
*From:* Don Dailey dailey@gmail.com
*To:* computer-go computer-go@computer-go.org
*Sent:* Wednesday, August 12, 2009 11:24 PM
*Subject:* Re: [computer-go] Dynamic komi at high handicaps
Terry,
I
On Fri, Aug 7, 2009 at 5:51 AM, Folkert van Heusden
folk...@vanheusden.comwrote:
What is superko?
My program keeps a list of all board-positions and then if it whants to
do a move it checks if the new board-position is in the list. If so, it
throws that move away. Are there other checks I
It's pretty easy to add this test to the client - there is an object
oriented package in the cgos code called gogame which follows a game and
reports any illegal moves.Just cut and paste it to client and when a new
game is started create a new game object, and use the object to verify the
Here are last few games of Pebbles where pebbles lost on time as black -
which is what would happen in a crash.
Pebbles is losing a lot of games on time.
794069|gnugo-3.7.12-l10F|1759|Pebbles|2155|2009-06-23
12:51|23130|306264|W+Time|y
796644|fuego-0.4-slow|2050|Pebbles|2144|2009-06-28
I only showed black games because you said pebbles was black.
- Don
On Sun, Aug 2, 2009 at 1:27 PM, Christoph Birk b...@ociw.edu wrote:
On Aug 2, 2009, at 8:05 AM, Don Dailey wrote:
Here are last few games of Pebbles where pebbles lost on time as black -
which is what would happen
I think the problem is yours, there is no known problem anything like this
in Ggogui and I have been using it for a long time.
Is your notation correct? In Go there is no 'i' file, we go from h to j
skipping i.You may already know that of course if you are a go player
but it still leads to
How is the center point handled?I assume it plays to the center point as
black and with either color it just ignores the center point in the symetry
calculations, right? So if it's playing white, symmetry is broken as
soon as white plays to the center because it cannot play a move that
2009/7/22 Andrés Domínguez andres...@gmail.com
2009/7/20 Stefan Kaitschick stefan.kaitsch...@hamburg.de:
Ofcourse they can know. They just have to check for it.
Those programs that do well against mirror go probably all do check for
it.
I think a strong MCTS could find the lines that make
It could be a matter of style as you say, not a matter of strength.My
main questions is whether it's been established as true that Zen really
plays poorly and Many Faces is brilliant against mirror go.Or does it
just seem that way based on casual observation?
The only reason I make an
I thought you played mirror go as white?
I'm not a go player, but it seems like it would be hard to win if you had
the white pieces with 0.5 komi and black mirrored everything you did.You
essentially start from a losing position anyway, right?
Does the human play to the center on the first
will get the make the capture
first, you could win by making sure this capture spoiled your opponent
mirror capture.
Is that the basic strategy or is there more?
- Don
On Mon, Jul 20, 2009 at 9:14 AM, Ingo Althöfer 3-hirn-ver...@gmx.dewrote:
Don Dailey wrote:
I thought you played mirror go
20, 2009 at 9:48 AM, Seo Sanghyeon sanx...@gmail.com wrote:
2009/7/20 Don Dailey dailey@gmail.com:
Again, I don't understand go so well, but how do you win against mirror
go?
You setup two ladders that collide.
--
Seo Sanghyeon
___
computer
On Wed, Jul 15, 2009 at 9:41 AM, Carter Cheng carter_ch...@yahoo.comwrote:
Where can I find information on these bridging protocols or are libraries
provided for this (to the 9x9 19x19 servers)?
The CGOS protocol is pretty easy to decode from the cgos client script which
is written in TCL.
I think you could do this with a binary tree - at each node keep a total of
the weight values of the subtree below the node.
If the pattern was hashed, then each bit could define a branch of the tree,
0 = left branch 1 = right branch.
Then you have a very simple divide and conquer algorithm.
In the loop i is always zero. I think your code is wrong.
You probably meant to loop over all the weights (or I should say on average
half the weights), and this code is slow if there are a lot of weights.
2009/7/16 Peter Drake dr...@lclark.edu
I must be missing something. Isn't the obvious
On Wed, Jul 15, 2009 at 11:37 PM, David Fotland fotl...@smart-games.comwrote:
So many complex ideas :) Why not just multiply the weight of each pattern
by a random number and pick the biggest result?
This is fine if you are looking for the slowest algorithm you can find.
But it does have the
On Tue, Jul 14, 2009 at 4:07 AM, Ingo Althöfer 3-hirn-ver...@gmx.dewrote:
Darren Cook wrote:
Ingo's suggestion (of two buttons to increment/decrement komi by one
point) was to make it easy for strong humans to test out the idea for us.
Don Dailey wrote:
There is no question that if you
There has been quite a few descriptions on this forum about how people do
this.
I am guessing, but I think most of the authors allocate a pool of memory and
manage this themselves.Are you writing in C?
In C you can declare a fixed size record (called a struct) and just make an
array of
which appear to be the same but differ
in crucial ways.
Thanks everyone for the help.
--- On Tue, 7/14/09, Don Dailey dailey@gmail.com wrote:
From: Don Dailey dailey@gmail.com
Subject: Re: [computer-go] memory management for search trees(basic
question)
To: computer-go computer
On Sun, Jul 12, 2009 at 8:07 AM, Benjamin Teuber benjamin.teu...@web.dewrote:
Hi,
I would like to know what exact experiments with virtual komi have
been made and why thay failed. To me, this idea seems very natural, as
it encodes the confidence of the stronger player that the weaker one
On Sun, Jul 12, 2009 at 8:59 AM, Benjamin Teuber benjamin.teu...@web.dewrote:
You just hit the nail on the head. Dynamic komi does not encourage a
program to overplay the position. Since you are starting from a losing
position you HAVE to overplay a bit. You have to attack when it is
2009/7/12 David Fotland fotl...@smart-games.com
e) use a knowledge system that knows what good moves look to prune or
bias the moves when way ahead or way behind. This is what many Faces does.
This is what I believe to be the most reasonable approach.
- Don
David
*From:*
-analsysis to figure out what needs to be done,
and by then you may already know what to do anyway and you have a more
convential program.
- Don
Christian
2009/7/12 Don Dailey dailey@gmail.com:
On Sun, Jul 12, 2009 at 8:59 AM, Benjamin Teuber benjamin.teu...@web.de
wrote:
You
On Sun, Jul 12, 2009 at 3:08 PM, Benjamin Teuber benjamin.teu...@web.dewrote:
It's not up to me to prove anything. It's up to you.
You entered a discussion in which you gave arguments (that I believe
are nonsense) ...
but at least fits the observation that this method does not work.
On Sun, Jul 12, 2009 at 7:12 PM, Matthew Woodcraft
matt...@woodcraft.me.ukwrote:
Don Dailey wrote:
I did try this myself but I don't have any data to show what I did.
What
I remember is that it's incredibly tricky - how do you actually know when
and how much to adjust? If the score
On Sun, Jul 12, 2009 at 8:10 PM, Darren Cook dar...@dcook.org wrote:
I would like to know what exact experiments with virtual komi
have been made and why thay failed. ...
I'm only aware of Don's experiment [1], which he admits he doesn't have
any details for and only remembers: I did a
I think we should open up to other ideas, not just dynamic komi
modification. In fact that has not proved to be a very fruitful technique
and I don't understand the fascination with it.
First we identify what it is we are trying to accomplish. You mentioned
improving the strength of MCTS go
On Sat, Jul 11, 2009 at 4:54 PM, Dave Dyer dd...@real-me.net wrote:
If you are in a lost position, good play is play that maximizes
the probability of a turnaround, which is quite different depending
on how far behind you are, and for what reason.
What maximizes the probability of a
On Mon, Jun 8, 2009 at 7:35 AM, Stefan Kaitschick
stefan.kaitsch...@hamburg.de wrote:
Thinking about why... In a given board position moves can be grouped
into sets: the set of correct moves, the set of 1pt mistakes, 2pt
mistakes, etc. Let's assume each side has roughly the same number of
On Tue, Jul 7, 2009 at 3:49 AM, Magnus Persson magnus.pers...@phmp.sewrote:
Quoting Oliver Lewis ojfle...@gmail.com:
Others on this list have reported in the past that the randomness is
actually very important. Playouts that are very heavy, no matter how
clever they are, actually reduce
On Tue, Jul 7, 2009 at 6:31 AM, Christian Nentwich
christ...@modeltwozero.com wrote:
The problem I think is to find a good tradeoff between heavyness and
speed. In my test with Valkyria vs Fuego, Valkyria is superior when the
number of playouts are the same. But Fuego can play 5 times more
1 - 100 of 1773 matches
Mail list logo
