Also, when it comes to literature, this is the place to start:
http://www.citeulike.org/group/5884/library
Peter Drake
http://www.lclark.edu/~drake/
On May 3, 2010, at 2:01 PM, Tobias Pfeiffer wrote:
Dear Computer-Go mailing-list subscribers,
I have a little question and I would be very
was the latest
large development but I suspect things have move beyond this now.
http://www.citeulike.org/group/5884/library
Peter Drake
http://www.lclark.edu/~drake/
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman
could conceivably cause the program some moves that
are illegal, but I believe (without strong evidence) that such events
are rare enough to ignore.
Peter Drake
http://www.lclark.edu/~drake/
On May 9, 2010, at 11:22 AM, Mark Boon wrote:
So nobody has made it work yet? That makes me feel
somewhere?
Peter Drake
http://www.lclark.edu/~drake/
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Yes, that explains it. Thanks!
Peter Drake
http://www.lclark.edu/~drake/
On Dec 3, 2010, at 4:03 PM, Petr Baudis wrote:
Hi!
On Fri, Dec 03, 2010 at 01:07:16PM -0800, Peter Drake wrote:
After playing many consecutive 9x9 games, Orego12 was cut off
because (apparently) someone tried
protocol:
http://www.lysator.liu.se/~gunnar/gtp/
Tournaments are held monthly on the KGS Go Server. Watch this list for
details.
If the conditions are right, I might want to try Go-programming in
near
future...
Joona Kiiski
Best of luck!
Peter Drake
http://www.lclark.edu/~drake
should find it particularly useful.
I'll be happy to answer any questions about setting it up and using
Orego.
Peter Drake
http://www.lclark.edu/~drake/
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman
I thought it already was one. What is the obstacle?
(I don't have a spare machine to run it on, but anyone else is welcome
to do so.)
Peter Drake
http://www.lclark.edu/~drake/
On Jan 10, 2011, at 1:07 PM, Brian Sheppard wrote:
Is it now possible to make Orego into a standard opponent
Not a problem.
Peter Drake
http://www.lclark.edu/~drake/
On Jan 10, 2011, at 5:41 PM, Brian Sheppard wrote:
Maybe I confused you with the term standard. There is a page
devoted to engines that meet certain criteria: http://computergo.wikispaces.com/CGOS+Standard+Engine+Packs
Orego
Yes, the Power of Forgetting paper.
Peter Drake
http://www.lclark.edu/~drake/
On Jan 11, 2011, at 3:47 AM, Aja wrote:
I discover that this upcoming paper is already available in your
website http://legacy.lclark.edu/~drake/Orego.html
I will try this interesting idea in Erica.
Aja
to
Java in this version so that new researchers wouldn't have to learn
another language.
Peter Drake
http://www.lclark.edu/~drake/
On Jan 11, 2011, at 9:41 AM, Joona Kiiski wrote:
Hi everyone,
During the last week I've been examining sources of different open
source go-engines (fuego, pachi
All true. We were hoping to get an easy boost by extracting joseki
from recorded games of strong players. Finding refutations of bad
moves is more difficult, because strong players don't make the bad
moves and weak players don't know how to refute them. :-)
Peter Drake
http
. StoneGrid had played
normally in the previous tests, so I guess it dislikes non-standard
board sizes.
You are correct about my location, but I don't understand your
reasoning. Also, I am the author of Orego, not StoneGrid.
Also, Orego has to be recompiled to change board sizes.
Peter
encourage other authors to try it.
It's reassuring to see that you got some strength improvement out of it!
Thanks,
Peter Drake
http://www.lclark.edu/~drake/
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman
I'm all for a learning policy, if you can figure out how to do it. :-)
Peter Drake
http://www.lclark.edu/~drake/
On Jan 25, 2011, at 11:31 AM, Aja wrote:
Hi Professor Drake,
I will try with more playouts. Thanks for your reminding.
I give an example to show my view: default policy should
Is this the correct way to state those time settings?
rules.time=19:00+10/0:30
Peter Drake
http://www.lclark.edu/~drake/
On Mar 5, 2011, at 12:47 AM, Nick Wedd wrote:
Reminder - it's tomorrow.
In message baht9kbxphanf...@maproom.demon.co.uk, Nick Wedd n...@maproom.co.uk
writes
That did the trick -- thanks for the instant response!
Peter Drake
http://www.lclark.edu/~drake/
On Jun 4, 2011, at 6:06 PM, Francois van Niekerk wrote:
You seem to be running an old version of kgsGtp. Try upgrading to
3.5.0 from http://www.gokgs.com/download.jsp
--
Francois van Niekerk
policy (used for the
first 10 moves of each playout) is to choose the (legal) move that the
classifier rates highest.
Is that the only use of the classifier in the system?
The above is the only use of the classifier.
Peter Drake
http://www.lclark.edu/~drake
to
one or three. (We were surprised by the latter result; we thought
deeper history might slow things down, but we didn't expect it to harm
performance.)
Peter Drake
http://www.lclark.edu/~drake/
___
Computer-go mailing list
Computer-go@dvandva.org
at these levels)
Yes -- as stated, this new method is not (yet?) competitive with
cutting-edge MCTS (e.g., RAVE and fancy domain-specific playouts). Our
claim is merely that it beats vanilla UCT.
Peter Drake
http://www.lclark.edu/~drake/
___
Computer-go
a result that
is better. (But maybe not much better; hard to tell because there is
no comparison to Orego+RAVE.)
Yes, RAVE is much better. :-)
Peter Drake
http://www.lclark.edu/~drake/
___
Computer-go mailing list
Computer-go@dvandva.org
http
is not magic, and cannot detect
patterns faster than the underlying statistics allow.
Of course.
Peter Drake
http://www.lclark.edu/~drake/
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
recent
moves. Again, no improvement.
Of course, it's possible that one of these ideas is valid and we just
did it wrong. We welcome experiments by others!
Thanks,
Peter Drake
http://www.lclark.edu/~drake/
___
Computer-go mailing list
Computer
True. It performed better at the times we tried, and vanilla MCTS did
not appear close to catching up. In the theoretical limit, though,
MCTS is clearly a richer representation.
On Jun 30, 2011, at 7:35 AM, Erik van der Werf wrote:
On Wed, Jun 29, 2011 at 10:26 PM, Peter Drake dr
add
patterns, atari, etc.
Disadvantages:
1) It's not immediately clear how to pre-initialize the tree to take
advantage of domain knowledge. Ideally this could be done through self-
play or examining recorded games.
Comments?
Peter Drake
http://www.lclark.edu/~drake
anyone else run into this?
Peter Drake
http://www.lclark.edu/~drake/
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
If you prefer to work in Java, Orego has (IMHO) clearly-written code,
including the features you mention:
http://legacy.lclark.edu/~drake/Orego.html
Peter Drake
http://www.lclark.edu/~drake/
On Aug 8, 2011, at 12:15 AM, Carter Cheng wrote:
Hi,
I am looking for some kind of baseline
moves, and was also won by oakfoam.
but none of the diagrams are labeled as involving Orego12. I presume
it's the one labeled Fuego vs pachi2 and captioned Movbe 49.
Thanks,
Peter Drake
http://www.lclark.edu/~drake/
On Aug 14, 2011, at 2:23 PM, Nick Wedd wrote:
Congratulations to pachi2
server (KGS, IGS, etc.).
Thanks,
Peter Drake
http://www.lclark.edu/~drake/
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
So far it seems to work well, but I don't see a way to start new games using
the app; you have to start them using the web interface to DGS, but then can
play using the app. (The New Game button is for recording a game.)
Thanks for all the feedback!
Peter Drake
http://www.lclark.edu/~drake
than Chess, but surely not THAT much.
Peter Drake
http://www.lclark.edu/~drake/
On Dec 13, 2011, at 5:35 PM, Jeff Nowakowski wrote:
On 12/13/2011 07:57 PM, Peter Drake wrote:
An exercise for the combinators and combinatrices out there:
How many different 2x2 Go games are there?
An unnamed
Is anyone interested in organizing a computer Go event at this year's US Go
Congress? I won't be able to make it this year.
Peter Drake
https://sites.google.com/site/drpeterdrake/home
___
Computer-go mailing list
Computer-go@dvandva.org
http
:
http://www.willamette.edu/cla/math/REU-RET/projects2012.html
Peter Drake
https://sites.google.com/site/drpeterdrake/home
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Hmm. Sounds like Orego never started. Try opening the GTP shell in GoGui.
Perhaps there's a useful error message there.
Peter Drake
https://sites.google.com/site/drpeterdrake/home
On Feb 10, 2012, at 11:26 AM, Ray Tayek wrote:
At 10:20 AM 2/10/2012, you wrote:
You might, for example, use
Hear, hear!
When it is kept up-to-date, this is an ENORMOUSLY valuable resource.
Peter Drake
https://sites.google.com/site/drpeterdrake/home
On Apr 21, 2012, at 6:27 AM, Rémi Coulom wrote:
Hi Thomas,
I cc to the whole list, because that may be of interest to everybody
@ and .).
Nick
--
Nick Wedd
n...@maproom.co.uk
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Peter Drake
http://www.lclark.edu/~drake/
___
Computer
have not yet read this
thesis.
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
, not 3/5.
BTW, I found this message in the archive:
http://www.mail-archive.com/computer-go@dvandva.org/msg03973.html
Disturbingly, I can't find it in my old email.
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
___
Computer-go mailing
I've got students trying to sign up for this mailing list. At least
one of them has not received any response for several weeks;
apparently the moderator has to approve these things. Does anyone know
a way around this or, alternately, an appropriate way to nudge the
moderator?
Thanks,
--
Peter
Nope, I'm using gmail and successfully receiving messages, while my
student is not.
On Tue, Jul 10, 2012 at 11:48 AM, Ingo Althöfer 3-hirn-ver...@gmx.de wrote:
Michael Alford wrote:
On 7/10/12 10:27 AM, Peter Drake wrote:
I've got students trying to sign up for this mailing list. At least
one
...but this led us to the solution. The confirmation email had been
classified as spam. He should be able to get on the list now.
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
On Tue, Jul 10, 2012 at 1:59 PM, Peter Drake dr...@lclark.edu wrote:
Nope, I'm using gmail
at anything. We now think it is more than 100 000 patterns.”
http://news.csu.edu.au//director/latestnews.cfm?itemID=C712563AEEA8398B7655D520C442F510printtemplate=release
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
___
Computer-go mailing list
Orego doesn't do this directly. Instead, we massing passing a possible
move, but with a very low prior success rate (1 win in 10 runs). It
will be chosen if all other moves lead to losses.
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
On Tue, Aug 7, 2012 at 1:34 PM, Michael
Orego 7.13 has been posted to the usual place:
https://sites.google.com/a/lclark.edu/drake/research/orego
We're also planning to put Orego up on GitHub in the very near future.
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
___
Computer
for forwarding this email to interested students!
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
New Image 3.jpg___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
, which is in Java:
https://sites.google.com/a/lclark.edu/drake/research/orego
The scoring methods are in the orego.core.Board class.
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
___
Computer-go mailing list
Computer-go@dvandva.org
http
We're trying to add this feature to Orego. We normally test against GNU Go.
If tell GNU Go komi=7, will it behave properly? (I can't tell from the
documentation.)
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
___
Computer-go mailing list
forward as appropriate.
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
/cgi-bin/mailman/listinfo/computer-go
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
I've got Orego's new account Orego60 (for using our entire 60-core cluster)
authorized as a ranked robot, but it's still playing free games. I logged
into the account and checked the rank box. What am I missing?
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake
a bot and do not talk.
gameNotes=AyaMC with 600sim
rules=chinese
rules.boardSize=19
rules.time=10:00+5x0:30
reconnect=t
undo=f
--**-
Regard,
Hiroshi Yamashita
- Original Message - From: Peter Drake dr...@lclark.edu
I already did that. What else?
On Sat, May 11, 2013 at 9:56 PM, Aja Huang ajahu...@gmail.com wrote:
Probably you have to login as Orego60 and check *Rank* in the profile.
Aja
2013/5/12 Peter Drake dr...@lclark.edu
I can't figure out what's different. Below is my configuration file
It looks like for some reason the current version of Orego is not
supporting final_status_list. That seems to be the issue. I'll get on
fixing that soon.
On Sat, May 11, 2013 at 10:12 PM, Peter Drake dr...@lclark.edu wrote:
I already did that. What else?
On Sat, May 11, 2013 at 9:56 PM, Aja
. As a workaround, we don’t really need
to clear the board when we get kgs-game_over, because KGS sends
clear_board at the beginning of the game.)
Any help would be appreciated.
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
___
Computer-go
Orego is now more thoroughly documented and open source on GitHub:
https://sites.google.com/a/lclark.edu/drake/research/orego
Feedback is appreciated!
We hope to make a number of improvements this summer.
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake
...and we just put up a version 7.15 (with better handling of dead stones
at the end of the game).
On Sun, Jun 30, 2013 at 8:01 AM, Martin Møller Skarbiniks Pedersen
traxpla...@gmail.com wrote:
Den 27/06/2013 22.35 skrev Peter Drake dr...@lclark.edu:
Orego is now more thoroughly
Is anyone else here going to the US Go Congress in Tacoma next week?
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
We had no amazing breakthroughs this summer, but we did explore some
interesting leads, which are presented in four posters found here:
https://sites.google.com/a/lclark.edu/drake/research/orego/publications-and-presentations
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake
consistent. This looks like 300 playouts
total, not per option.
In either case, my program (Orego) clearly needs MUCH heavier playouts. :-)
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
___
Computer-go mailing list
Computer-go@dvandva.org
, “Intelligent Agents for the Game of Go,” IEEE
Comput. Intell. Mag., vol. 5, no. 4, pp. 28–42, 2010.
Regards,
Cameron
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
--
Peter Drake
https
...@gmail.com .
Nick
--
Nick Wedd
n...@maproom.co.uk
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake
A naive question:
In what situations is it better to use Coulom's Elo method vs his CLOP
method for setting parameters? It seems they are both techniques for
optimizing a high-dimensional, noisy function.
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake
On Tue, Feb 11, 2014 at 1:08 PM, Petr Baudis pa...@ucw.cz wrote:
Hi!
On Tue, Feb 11, 2014 at 11:42:24AM -0800, Peter Drake wrote:
A naive question:
In what situations is it better to use Coulom's Elo method vs his CLOP
method for setting parameters? It seems they are both techniques
require that dead stones are
removed, and I can't figure out how to do that easily. Or do you just
assume everything on the board is alive?
One reasonable approach, which Orego uses, is to do a bunch of extra Monte
Carlo playouts. If a group survives often enough, consider it alive.
--
Peter
/mailman/listinfo/computer-go
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
overwhelming everyone
with technical details most of the discussion will take place on the Orego
list. If you're interested in participating (or lurking) in these
discussions, email me (dr...@lclark.edu) and I can add you to the list.
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake
this mean that you maintain the bitmaps in parallel with the other
structures?
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
;
(The rest of what you describe sounds like what I'm doing, which is
reassuring.)
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
?
They're *almost* equivalent to AGA rules, but AGA uses situational superko
while Chinese uses positional superko. The difference would only come up
extremely rarely, potentially making for a nasty bug.
For those who don't know these terms:
http://senseis.xmp.net/?Superko
--
Peter Drake
https
-bin/mailman/listinfo/computer-go
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
be a good chance to
check memory and time control.
Regards,
Hiroshi Yamashita
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
--
Peter Drake
https://sites.google.com/a/lclark.edu
://senseis.xmp.net/?KGSBotRatings (only goes back to 2007)
http://www.computer-go.info/h-c/ (a bit difficult to condense into a single
line graph)
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
___
Computer-go mailing list
Computer-go@dvandva.org
http
anything conventional. Later in 2007 SlugGo beat an 8k human
at the Cotsen Open running on an 8 core Xenon Mac Pro tower.
I hope this helps.
Cheers,
David G Doshay
ddos...@mac.com
On 26, Sep 2014, at 4:48 PM, Peter Drake dr...@lclark.edu wrote:
I'm giving a talk on computer Go (and how
eyes.
I *think* this definition works regardless of whether adjacent blocks have
one or two eyes. (If filling is always illegal, it certainly requires
capturing all of the surrounding stones.)
Is there a horrible flaw in my definition?
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake
On Thu, Oct 23, 2014 at 3:38 PM, Robert Jasiek jas...@snafu.de wrote:
On 24.10.2014 00:17, Peter Drake wrote:
An eye is a vacant point, surrounded [orthogonally] by stones of the same
color, that can only be filled by simultaneously capturing all of the
surrounding stones.
If it can
On Thu, Oct 23, 2014 at 11:36 PM, Robert Jasiek jas...@snafu.de wrote:
On 24.10.2014 01:19, Peter Drake wrote:
Isn't A an eye here?
.
.xxx.
.xAx.
.xxx.
.
(Remember, I'm trying to define eye, not life).
If you want to use eye in a broader sense, then every region
Well, yes, of course. But you can still have ONE eye in a group that is not
alive.
On Fri, Oct 24, 2014 at 9:54 AM, Robert Jasiek jas...@snafu.de wrote:
On 24.10.2014 17:40, Peter Drake wrote:
Don't you need TWO eyes to be alive [...]?
It is possible for one big eye(space) to transform
.
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
___
Computer-go mailing list
Computer-go@dvandva.org
http
in all directions (including diagonally) by
stones of one color, which is a sufficient but not a necessary criterion.
On Fri, Oct 24, 2014 at 1:07 PM, Robert Jasiek jas...@snafu.de wrote:
On 24.10.2014 21:47, Peter Drake wrote:
you can still have ONE eye in a group that is not alive.
This is bad
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi
___
Computer-go mailing list
Computer-go@dvandva.org
http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
--
Peter Drake
https://sites.google.com/a/lclark.edu/drake/
___
Computer-go mailing list
Computer-go
81 matches
Mail list logo