[computer-go] CGOS down?
Hi, It seems that CGOS 9x9 is down: it only asks the name, password and command list to the engine. It says on the web interface that games are playing, but this webpage is not refreshing. Cheers, Guillaume ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
[computer-go] MoGo beats pro: The website
Dear all, There were details that were unclear about the victory of MoGo. Hence I created a website to gather useful information about this game: http://www.cs.unimaas.nl/g.chaslot/muyungwan-mogo/ Cheers, Guillaume Message d'origine De: [EMAIL PROTECTED] de la part de Sylvain Gelly Date: mer. 8/13/2008 19:54 À: computer-go Objet : Re: [computer-go] What was the specific design of the Mogo versionwhich beat the pro... C++ on linux (with a port on windows using cygwin libraries for the binary release) Sylvain 2008/8/13 steve uurtamo [EMAIL PROTECTED] And what language/platform is Mogo written in; C/C++, Java, Assembly, PHP, etc.? This made coffee spray out of my nose (PHP). I think that C is most likely, based upon how they parallelized it. Did you read the list posting that mentioned (briefly) how they scaled it up? s. ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/ winmail.dat___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
[computer-go] Mogo beats pro: the hardware
Dear all, The machine that was used by MoGo yesterday is the Dutch supercomputer Huygens, situated in Amsterdam. Huygens was provided by SARA (www.sara.nl) and NCF(http://www.nwo.nl/nwohome.nsf/pages/ACPP_4X6R5C_Eng). Huygens was upgraded on August 1 to 60 Teraflops (Peak), so porting MoGo with this short notice for the match was a lot of hard work and stress. But the result showed it was worth it! Huygens is constituted of 104 nodes of 16 dual-cores POWER6 processors at 4.7 GHz each (with 128G of RAM). MoGo was using 25 nodes, i.e., 800 cores and nearly 15 Teraflops. By comparison, Deep Blue was using only 11 Gigaflops. The structure of Huygens with powerful processors and numerous cores per node is ideal for MoGo. It would be less efficient to use a supercomputer with few cores per node, e.g., Blue Gene. The parallelization was performed using Pthreads and OpenMPI. On each node, two search-trees were built, each one using 32 threads. Thanks to the SMT technology, it is actually more efficient to use two threads per core. Indeed, while one thread is looking in memory, an other thread can use the core. Every 350 milliseconds, the nodes were communicating their tree to one an other using OpenMPI. Finally, we would like to help all the people who helped us to port the code, both from the french INRIA/CNRS and the dutch NCF/SARA. Cheers, and see you in Beijing! The Mogo Team: http://www.lri.fr/~teytaud/mogo.html PS: A nice picture of Huygens, and further information, can be found here: www.sara.nl winmail.dat___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
[computer-go] Call for participation: Next Computer Olympiad in Beijing
Dear Go programmers, The 13th Computer Olympiad (CO) will be held from September 28 to October 5 in Beijing, China. This event will be held together with the Conference on Computers and Games 2008 (CG 2008) (September 29 - October 1), and the 16th World Computer-Chess Championship (WCCC) (September 28 - October 5) Location: the Beijing Golden Century Golf club, Fangshan, Beijing, China. For more information: http://www.grappa.univ-lille3.fr/icga/event_info.php?id=18 I hope to meet you there, Guillaume ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
RE: [computer-go] More UCT / Monte-Carlo questions (Effect of rave)
I also implemented RAVE in Mango. There was a few points of improvements (around 60 Elo points with gnugo as reference), but as much as in the paper of Gelly and Silver :( (around 250 Elo points if I remember well) It might be that the effect of RAVE depends a lot on the simulation strategy. Indeed, sometimes my RAVE was playing very good moves but also very bad ones. Guillaume -Original Message- From: [EMAIL PROTECTED] on behalf of Magnus Persson Sent: Wed 06/02/2008 00:42 To: computer-go@computer-go.org Subject: Re: [computer-go] More UCT / Monte-Carlo questions Quoting Gunnar Farnebäck [EMAIL PROTECTED]: I have never managed to implement RAVE successfully. It made my program significantly slower but no stronger even at a fixed number of simulations. I get a small effect from RAVE, My rationalisation is that if the program is rich with other features to improve performance RAVE may not add that much. -Magnus ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/ winmail.dat___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
RE: [computer-go] visual tree (gogui extension) help request
Hi, You have to put the file vtbeta.jar in the lib directory of the gogui. For me, it is ~/Go/gogui-0.9.2/lib. vtbeta.jar uses the other files that are in this directory. Note that the visual tree is not compatible with gogui 1.x yet, only with version 0.9.x :( Visual Tree was also used by Sylvain Gelly in Mogo. I'm happy that other people are trying it, even if I don't have time to work on it now (for instance I would like to upgrade it for the gogui 1.x). However there are all the sources plus a lot of documentation on my website, so anybody can adapt it to his requirements. Best, Guillaume -Original Message- From: [EMAIL PROTECTED] on behalf of Peter Christopher Sent: Sat 12/01/2008 11:14 To: computer-go@computer-go.org Subject: [computer-go] visual tree (gogui extension) help request Hi, By any chance does anyone have a tip to help me get VisualTree working? http://www.cs.unimaas.nl/g.chaslot/vt/vt09/docs/install.html says I should run java -jar vtbeta.jar but this produces the error: Exception in thread main java.lang.NoClassDefFoundError: net/sf/gogui/utils/ErrorMessage The original page also states to store the jar in a certain lib directory I cannot find on my linux ubuntu system. Gogui does work fine. I use it regularly (great tool). Thanks in advance. Peter ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/ winmail.dat___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
RE: [computer-go] Re: Binary release of MoGo
00:00 1 name tried to open opening, success 0-- in grey Here it does not find the file, because the file is with the binaries and gogui (at least your version) looks into the gogui/bin directory. But that does not prevent MoGo to work. I think gogui is in fact looking for files in the directory from which it is launched. Try this to copy the opening database in this directory. Thanks a lot Sylvain! It runs perfectly on an Opteron 2.6GHz. But not on a Power5+ processor. (by the way a Power5+ processor 1.9Ghz is at least 25 percent slower than an Opteron 2.6GHz, for my program...) Guillaume ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
[computer-go] Results of 19x19 Go in the 12th Computer Olympiad
Congratulation to Mogo for winning the gold medal with a perfect score of 7 out of 7. CrazyStone received the silver medal and GnuGo the bronze medal. The final match between CrazyStone and Mogo, was commented live on KGS by Guo Juan (5p). Guo Juan played several fast games against Mogo: - one 19x19 game without handicap. - one 19x19 game with 9 stones handicap. - three 9x9 games. It can be noticed that Mogo was able to win one of the 9x9 games. The commentary and the games, followed by more than 380 go players, can be found on KGS in the list of games of GuoJuan. Full results of this tournament can be found here: http://www.grappa.univ-lille3.fr/icga/tournament.php?id=167 All the games of the tournaments are available here: http://www.grappa.univ-lille3.fr/icga/round.php?tournament=167 In the mean time, Rybka won the World Computer Chess Championship. (http://amsterdam2007.icga.org) These tournaments were organized by the ICGA (International Computer Games Association). Best regards, and looking forward to the next Computer Olympiad (Beijin 2008), Guillaume winmail.dat___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
[computer-go] Professional Go player at the Computer Olympiad
Dear all, We are proud to announce that Guo Juan, 5d professional from China, will be present at the Computer Olympiad in Amsterdam on the Sunday 17th of June. She will play against the winner of the Olympiad and comment the most interesting games. We will broadcast as many Go games and comments as possible on KGS. The website of the Olympiad can be found here: http://amsterdam2007.icga.org/ Best regards, and hopping to meet some of you there, Guillaume ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
[computer-go] JCIS extended abstract
Dear all, Following the example of Rémi, I would like to share with you a paper that I wrote which describe some important elements of my Go program Mango. I submitted this paper recently to the JCIS workshop 2007. Due to the fact that it was an extended abstract, a lot of details are missing. I am now working on the full paper, which will contain much more information. The extended abstract can be found here: www.cs.unimaas.nl/g.chaslot/papers/pMCTS.pdf Cheers, Guillaume PS: To answer's Sylvain question: Modifying the probability distribution of moves in the tree was more effective in Mango than the improvement done on the simulation strategy. However, Mango simulation strategy is not as good as CrazyStone's or Mogo's one. So I guess Mango won against Mogo in the last KGS tournament because it had a better way to use domain knowledge in the Tree. ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
RE: [computer-go] JCIS extended abstract
How do you handle n_i = 0 in equation 1.2? I don't compute it when n_i is not equal to 1. At that point, I give every node a high upper bound in order to visit every node once. Remember that I am using progressive unpruning in the same time, so only the best nodes according to the heuristic are visited. I'd assume that if a heuristic says a move is really bad (H_B = 0), then you'd want to avoid simulations for a while. Yes! This is exactly the purpose of progressive unprunning. p_hat = (w_i + n_h*H_B)/(n_i+n_h) Interesting... But then how do you compute n_h in practice If I understood Remi correctly, he computes the value by adding one virtual loss and one virtual win for each gamma, ie, with your notation: p_hat= (w_i+gamma)/(gamma*2+n_i) This formula tends to center the probability distribution on 0.5. But if a pattern is good, I would rather shift the distribution towards 1! Adding a term in gamma/n_i as I do in my paper is a solution to shift the probability distribution to towards 1 which works well in practice. For the full version of my paper I will compare different ways to modify the probability distribution according to knowledge. I believe there is no optimal way to do that :( This problem is fundamental, because it gives the opportunity to include previously learned knowledge into Monte-Carlo Tree Search. -Original Message- From: [EMAIL PROTECTED] on behalf of Jason House Sent: Thu 17/05/2007 22:26 To: computer-go Subject: Re: [computer-go] JCIS extended abstract How do you handle n_i = 0 in equation 1.2? I'd assume that if a heuristic says a move is really bad (H_B = 0), then you'd want to avoid simulations for a while. Also, has mango experimented with other related strategies for a soft transition? In my own mathematical recreation, I came up with a slightly variant method. Just to keep the text short, here's the notation that I'll use... N = number of simulations through parent node n_i = number of simulations through this node w_i = winning simulations through this node v_i = w_i / n_i H_B = heuristic_bias n_h = heuristic confidence (used in my equation) UCTdelta = sqrt(ln(N)/n_i) p_hat = estimated probability of winning with soft transition ... then equation 1.2 is p_hat + UCTdelta For you, p_hat = (w_i + H_B)/n_i I derived and posted the following formula to the list: p_hat = (w_i + n_h*H_B)/(n_i+n_h) The two equations are fairly similar... especially if n_h is set to 1. Then the only difference is that I replace n_i with (n_i+1)... and allows p_hat to be calculated with n_i = 0. I made no attempt to handle UCTdelta, but replacing n_i in there with n_i+n_h is probably reasonable On 5/17/07, Chaslot G (MICC) [EMAIL PROTECTED] wrote: Dear all, Following the example of Rémi, I would like to share with you a paper that I wrote which describe some important elements of my Go program Mango. I submitted this paper recently to the JCIS workshop 2007. Due to the fact that it was an extended abstract, a lot of details are missing. I am now working on the full paper, which will contain much more information. The extended abstract can be found here: www.cs.unimaas.nl/g.chaslot/papers/pMCTS.pdf Cheers, Guillaume PS: To answer's Sylvain question: Modifying the probability distribution of moves in the tree was more effective in Mango than the improvement done on the simulation strategy. However, Mango simulation strategy is not as good as CrazyStone's or Mogo's one. So I guess Mango won against Mogo in the last KGS tournament because it had a better way to use domain knowledge in the Tree. ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/ winmail.dat___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
[computer-go] 12th Computer Olympiad
Dear Go programmers, The ICGA has concluded the negotiation for organizing the WCCC 2007, the 12th Computer Olympiad, and an accompanying scientific workshop . The events will take place in Amsterdam, The Netherlands, 11-18, June 2007. The workshop will be held on Friday 15. - Sunday 16. June 2007. The Call for papers can be found on the website of the ICGA: http://www.icga.org/ An excursion is scheduled for all interested participants. A detailed schedule will be announced as soon as possible. The Computer Olympiad will include a 9x9 and a 19x19 Go tournament. More information will be available soon on the tournaments page: http://tournaments.icga.org/ Best Regards, Guillaume winmail.dat___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
RE: [computer-go] Useless moves in the endgame
Mango passes as soon as the opponent passes two times in a row. Might this lead to bugs in some situations? Anyway this is very nice for playing against humans and GnuGo. Guillaume -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Benjamin Teuber Sent: Tuesday, January 09, 2007 4:32 PM To: computer-go Subject: [computer-go] Useless moves in the endgame I just lost my first game against MoGo on KGS, 9x9, 0.5 komi, I was white. Impressing! But as a human, you don't like the useless endgame-moves MC-programs play against you when they know they win anyways. In order to make these programs more attractive for humans, I would like them to play the move winning by the biggest amount of points once several moves have the same high winning probability at the endgame. What do you think about this? ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/ ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/