Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
On Tue, Jan 03, 2012 at 08:42:31AM +0100, Robert Jasiek wrote: Reasonable? As long as a player does not know when he going to play (because he has to participate in the game match accepting click war), he suffers from the psychological disadvantage of suddenly being involved in a game. There's no very good solution. With some programs, you can install them on your computer and play them at their leisure, but the power of your hardware matters a lot. Humans make blunders in byoyomi only games. I do not know how many but it is quite some number. I also do not know how many blunders computers make. One thing I do know: In a real world game with long thinking times, the 5d+ human's blunder rate per game is below 1 move on average. IOW, you cannot compare online byoyomi games with human long thinking time games at all. What is blunder rate? When you watch a professional review of a high dan amateur game, there certainly does seem to be a lot of blunders. Isn't it a matter of perspective? :-) It is not as bad as Nihon Kiin certificates for programs Wow, did that ever happen?! but almost as bad to set computer-friendly conditions all the time. Do you have any precise idea in mind that would allow reasonable number of (strong) people to play a program, avoid clicking matches and be friendlier to the humans? Have the courage to compete under human conditions! Enter human tournaments! That's easy to say. Do you know a tournament where a program can enter? I have tried few times with Pachi, never successfully yet. I think some other authors tried as well in the past. Petr Pasky Baudis ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
On 03/01/2012 07:47, Jouni Valkonen wrote: On 3 January 2012 00:53, Aja Huang ajahu...@gmail.com mailto:ajahu...@gmail.com wrote: Blitz games such as 15s/move favor MCTS programs. I expect both Zen and CrazyStone will drop to 4d in longer games. It would be nice to have bot Zen and CS in slower (ca. 60min) tournaments. I remember that Zen has participated at least to one KGS tournament, and did good, but more data would be nice. from slower thinking time. Byouyomi playing is always little bit difficult for humans. The tournaments I organise on KGS do not include any with an hour each. This is because most of them are held within a single 8-hour session, and are Swiss, so an hour each would mean only for rounds, which I think is not enough. A few of them are held over most of a week, and have all had at least two hours per player per game. These tournaments all use something close to absolute time (Canadian overtime of 10 moves in 30 seconds) so as to use the time effectively. I am open to persuasion to change any of this. Zen has been taking a short holiday from these tournaments, and CrazyStone has not played in one since April 2010 http://www.weddslist.com/kgs/past/58/index.html when it came second, behind Zen. The next one will be on January 15th, with 19x19 boards and time limits of ~30 minutes each. I am hoping that Zen and CrazyStone will both take part. Nick Also, especially, I would love to see strong gobot playing in EGC 2012 main tournament. Of course there is lots of organizing thing to do, but those two evil gobots are good enough and still not yet too good to participate into serious human tournaments with long thinking times. I personally prefer to play 80×20sec at KGS. It quite nice playing pace. There is 28 minutes for thinking + 20 sec for each moves. It is significantly better for humans than 20 min + 5×30 sec, although total game is length is roughly the same. –Jouni ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go -- Nick Wedd n...@maproom.co.uk ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
On 3 January 2012 13:25, Petr Baudis pa...@ucw.cz wrote: On Tue, Jan 03, 2012 at 08:42:31AM +0100, Robert Jasiek wrote: but almost as bad to set computer-friendly conditions all the time. Do you have any precise idea in mind that would allow reasonable number of (strong) people to play a program, avoid clicking matches and be friendlier to the humans? yes there is very clear idea, that we want strong gobots into human tournaments. Especially into slow KGS tournaments and I think that this would be sufficient and objective enough source for data. These conditions are more friendly for humans. Although they are not perfect such as in EGC main tournament. –Jouni ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
The tournaments I organise on KGS do not include any with an hour each. it would be nice for machines to be allowed to enter the regular tournaments as well. :) or, if that's too hard to organize, to allow people to enter some of the machine tournaments. ;) s. ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
On 03/01/2012 11:25, Petr Baudis wrote: snip Have the courage to compete under human conditions! Enter human tournaments! That's easy to say. Do you know a tournament where a program can enter? I have tried few times with Pachi, never successfully yet. I think some other authors tried as well in the past. Many years ago, when running a Go tournament for humans, I allowed a program to enter. It only played in rounds in which the number of human entrants was odd, so as to avoid byes (I thought, surely playing a game against a program is better than sitting around reading a newspaper for a round?). But I was reprimanded by the British Go Association, and told never to do this again. Nick -- Nick Wedd n...@maproom.co.uk ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
i think that kgs players might be more open-minded about this. s. On Tue, Jan 3, 2012 at 3:44 AM, Nick Wedd n...@maproom.co.uk wrote: On 03/01/2012 11:25, Petr Baudis wrote: snip Have the courage to compete under human conditions! Enter human tournaments! That's easy to say. Do you know a tournament where a program can enter? I have tried few times with Pachi, never successfully yet. I think some other authors tried as well in the past. Many years ago, when running a Go tournament for humans, I allowed a program to enter. It only played in rounds in which the number of human entrants was odd, so as to avoid byes (I thought, surely playing a game against a program is better than sitting around reading a newspaper for a round?). But I was reprimanded by the British Go Association, and told never to do this again. Nick -- Nick Wedd n...@maproom.co.uk ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
On 03.01.2012 12:25, Petr Baudis wrote: What is blunder rate? The blunder rate is the average number of blunders per game. A blunder is a mistake that is a) big and b) the player could have avoided rather easily by thinking a bit more and given his go insight. When you watch a professional review of a high dan amateur game, there certainly does seem to be a lot of blunders. I am not speaking of ordinary mistakes (like choosing a wrong direction) but of blunders (like overlooking an atari in three moves). While the difference between ordinary mistake and blunder is hard to define, I can always identify either in my games because there is a quantum jump between the different levels of mistakes. Isn't it a matter of perspective? :-) No. It is a matter of proper usage of the phrases ordinary mistake and blunder. It is not as bad as Nihon Kiin certificates for programs Wow, did that ever happen?! I though it was like that when the programs were about 9k but given 3k certificates. Do you have any precise idea in mind that would allow reasonable number of (strong) people to play a program, avoid clicking matches and be friendlier to the humans? Play for the humans real world games. Ask players of appropriate strength. Assign match schedules. Do you know a tournament where a program can enter? I agree that it requires work. -- robert jasiek ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
On 03/01/2012 11:42, steve uurtamo wrote: The tournaments I organise on KGS do not include any with an hour each. it would be nice for machines to be allowed to enter the regular tournaments as well. :) Sometimes, they are. See http://www.gokgs.com/tournEntrants.jsp?sort=sid=600 http://www.gokgs.com/tournEntrants.jsp?sort=sid=601 Zen came second in both of these; pachi also did well. The decision here will probably be with KGS admin 'sweety'. Most regular tournaments on KGS are open only to paying customers, i.e. KGS+ subscribers. So an issue is likely to be - do these paying customers want bots playing in their tournaments? or, if that's too hard to organize, to allow people to enter some of the machine tournaments. ;) I can consider that. How would I select the human players? I can't let people join without restriction, that might produce many more human players than bots, and undermine a major source of KGS revenue. Though I guess the tough schedule, close to eight hours of solid play, would be a deterrent. Nick -- Nick Wedd n...@maproom.co.uk ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
I can consider that. How would I select the human players? I can't let people join without restriction, that might produce many more human players than bots, and undermine a major source of KGS revenue. Though I guess the tough schedule, close to eight hours of solid play, would be a deterrent. i dunno, the first 4 (8?) people to mail you from the list with kgs accounts? i'd be surprised if there were a flood of people willing to play that very tough schedule. or if there were, more than once. s. ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
Hello, my two cent: or, if that's too hard to organize, to allow people to enter some of the machine tournaments. ;) I can consider that. How would I select the human players? People may apply. Only players with several KGS games (for instance = 10 within the last 30 days before the application) against strong bots are acceptable. If there are still more applicants than slots those with the highest KGS ratings should be accepted. Ingo. -- NEU: FreePhone - 0ct/min Handyspartarif mit Geld-zurück-Garantie! Jetzt informieren: http://www.gmx.net/de/go/freephone ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
i'd remove or alter the second requirement, it's actually quite hard to get that many games against strong bots, and doesn't add much (anything?) to the play for people to have done so. what you would want to see is strong players in a tournament, right? i agree about the rank ordering. s. On Tue, Jan 3, 2012 at 4:14 AM, Ingo Althöfer 3-hirn-ver...@gmx.de wrote: Hello, my two cent: or, if that's too hard to organize, to allow people to enter some of the machine tournaments. ;) I can consider that. How would I select the human players? People may apply. Only players with several KGS games (for instance = 10 within the last 30 days before the application) against strong bots are acceptable. If there are still more applicants than slots those with the highest KGS ratings should be accepted. Ingo. -- NEU: FreePhone - 0ct/min Handyspartarif mit Geld-zurück-Garantie! Jetzt informieren: http://www.gmx.net/de/go/freephone ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
On Tue, Jan 3, 2012 at 2:42 AM, Robert Jasiek jas...@snafu.de wrote: On 02.01.2012 23:47, David Fotland wrote: 15 seconds is pretty reasonable for a quick game, and 9 periods allows a couple of long thinks. LOL. Long thinks means allowing for 30 minutes of difficult LD solving! reasonable for a quick game is too imprecise. It is only a special type of quick game: online byoyomi only style. E.g., real world sudden death quick games have a very different nature. Reasonable? As long as a player does not know when he going to play (because he has to participate in the game match accepting click war), he suffers from the psychological disadvantage of suddenly being involved in a game. Humans make blunders in byoyomi only games. I do not know how many but it is quite some number. I also do not know how many blunders computers make. One thing I do know: In a real world game with long thinking times, the 5d+ human's blunder rate per game is below 1 move on average. IOW, you cannot compare online byoyomi games with human long thinking time games at all. It is not as bad as Nihon Kiin certificates for programs but almost as bad to set computer-friendly conditions all the time. Have the courage to compete under human conditions! Enter human tournaments! I know byoyomi is traditional, but I believe the Fischer clock is far more sane for human play. I believe you could play the games with less time using Fischer with higher quality too. With Fischer the time you don't use is never taken from you and there is not the constant clock pressure. Even if the main time was 5 minutes with 5 seconds fischer increment I think it would be far more conducive to strong play than 15 seconds byoyomi. Do they use fischer clock at all in any go competitions? -- robert jasiek __**_ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/**mailman/listinfo/computer-gohttp://dvandva.org/cgi-bin/mailman/listinfo/computer-go ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
The American Go Association allows computer entrants. I found the following in the AGA tournament guide: C. Computer entry. Computers may enter tournaments under certain conditions: 1. Only the inventor of the hardware/program or his/her designated agent may enter the computer (hereafter, either inventor or agent are called the operator.); 2. The computer must correctly handle any move legal for it or its opponent to make andmust not make any illegal moves; 3. Both computer and operator must be AGA members; 4. The operator must play computer moves on a regular board and punch the clock for the computer; 5. The operator may enter or adjust playing parameters before a round begins, but not during a round; 6. The computer's clock must be left ticking if the operator must fix hardware or software problems. 7. The operator may offer to resign on the computer's behalf. D. Classes of computer participation. There are three classes of computer tournament participation. Tournament publicity should indicate what class a tournament is ahead of time; if not announced, the tournament is automatically class B. The TD should also announce the class of tournament before first round pairings. 1. Class A: no computer entrants allowed. 2. Class B: computers allowed, but humans have the right to refuse computer opponents. Humans wishing to do so must notify the TD before first round pairings. 3. Class C: computers allowed; humans may not refuse computer opponents. Terry McIntyre terrymcint...@yahoo.com Unix/Linux Systems Administration Taking time to do it right saves having to do it twice. ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
steve uurtamo wrote: i'd remove or alter the second requirement, it's actually quite hard to get that many games against strong bots, and doesn't add much (anything?) to the play for people to have done so. The main reason to have is to avoid that someone enters who has no (or almost no) experience with bot play and later starts complaining (or steps back silently) when the tournament does not run well. 10 games against strong bots within 30 days would be one possible condition; it might also be okay to ask instead for 5 games within the last 90 days. Ingo. PS: By strong bot I mean not only Zen and CrazyStone, but also Pachi, Aya, ManyFaces, GinseiIgo, Steenvreter, Gomorra ... what you would want to see is strong players in a tournament, right? i agree about the rank ordering. s. On Tue, Jan 3, 2012 at 4:14 AM, Ingo Althöfer 3-hirn-ver...@gmx.de wrote: Hello, my two cent: or, if that's too hard to organize, to allow people to enter some of the machine tournaments. ;) I can consider that. How would I select the human players? People may apply. Only players with several KGS games (for instance = 10 within the last 30 days before the application) against strong bots are acceptable. If there are still more applicants than slots those with the highest KGS ratings should be accepted. Ingo. -- NEU: FreePhone - 0ct/min Handyspartarif mit Geld-zurück-Garantie! Jetzt informieren: http://www.gmx.net/de/go/freephone ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go -- Empfehlen Sie GMX DSL Ihren Freunden und Bekannten und wir belohnen Sie mit bis zu 50,- Euro! https://freundschaftswerbung.gmx.de ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
Hello Peter, Have the courage to compete under human conditions! Enter human tournaments! That's easy to say. Do you know a tournament where a program can enter? I have tried few times with Pachi, never successfully yet. I think some other authors tried as well in the past. Years ago (in 2001, 2002, 2003; before the Monte Carlo revolution) we had human go tournaments in Jena where some human+bot-constructions (3-Hirn and others) participated. Three details made it a success: * Participants declared before round 1 if they were willing to play against such beasts. * We had prizes (little books) for those who actually played against them. * For the local and regional press (and TV) computer participation was a highlight. They mainly reported on these experiments, but also about the tournament itself. Ingo. Petr Pasky Baudis ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go -- Empfehlen Sie GMX DSL Ihren Freunden und Bekannten und wir belohnen Sie mit bis zu 50,- Euro! https://freundschaftswerbung.gmx.de ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
Hi David, David Fotland on CLOP-optimization: I tried it, but got no benefit so far. It claimed to find better settings for most parameters, but when I used them the program wasn’t any stronger. Interestant. Had it similar strength or did it even become weaker? How often did the move proposals by your older ManyFaces and the CLOP-MF differ? Ingo. -- NEU: FreePhone - 0ct/min Handyspartarif mit Geld-zurück-Garantie! Jetzt informieren: http://www.gmx.net/de/go/freephone ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
[Computer-go] CrazyStone in the 5-dan footsteps of Zen
Maybe a tournament is not the best way to see quality computer/human games. There are better ways to measure the computer/computer performance and the human/human performance is not interesting here. We could simply schedule computer/human games on KGS (e.g., 3 times a year, one in each time zone afternoon) with around 4 KGS 5d+ humans and the 2 bots Zen and CrazyStone. Obvious human candidates are: Aja Huang, Robert Jasiek, Stefan Kaitschick, BotHater (don't know his name). Humans could use this list to subscribe and the pairings could be listed in advance. I don't think there are masses of KGS 5d+ players. It would be fun to watch. Jacques. ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
On 03.01.2012 13:09, Ingo Althöfer wrote: The open tournament has 10 rounds A human championship is not suitable for computers near top ranks. Other EGC tournaments are an option more easily. In particular, CG tournaments or side events specifically for computers and humans. -- robert jasiek ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
An observation: When computers get strong enough to be a threat to the stronger players in these tournaments (which is probably already the case), tournaments will end up all being Class A. It will become a default. Even being called Class A (no computers) will make it the defacto standard.If you didn't know the difference would you rather tell your friends you were in a class A tournament or a class C tournament? They built some bias into the rules for this just by their naming convention. Of course it is their right to make the rules, this is just an observation. Years ago in computer chess there were very similar rules about when a computer could participate and it was basically up to the discretion of the organizer - but the default policy was NO, it had to be stated if they were allowed. Almost immediately that closed the doors on computers playing at almost every event. I think there are 2 or 3 improvements here to what we had in chess, such as the possibility of Class C tournaments.I also like that computers should not be given additional consideration such as time-outs for hardware issues - that only serves to antagonize people who don't care about computers and have to suffer the scheduling consequences.I also agree that computers should not win prize money. To be perfectly frank about this my own experience with bringing my own chess program to tournaments has been rather negative and I tend to side with the human players who come expecting to play other humans even if they don't (or forget) to say so.So it makes sense (to me) that there should only be class A and class C tournaments. Class C tournaments should be organized specifically for computer participation with a kind of equal status between computers and humans. Humans can be enticed to come to these with the right incentives. One incentive is that computers can win prize money but that a portion (or all) of it is distributed to whoever played the computer. Don The TD should also announce the class of tournament before first round pairings. 1. Class A: no computer entrants allowed. 2. Class B: computers allowed, but humans have the right to refuse computer opponents. Humans wishing to do so must notify the TD before first round pairings. 3. Class C: computers allowed; humans may not refuse computer opponents. Terry McIntyre terrymcint...@yahoo.com Unix/Linux Systems Administration Taking time to do it right saves having to do it twice. ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
On 03.01.2012 13:58, Ingo Althöfer wrote: 10 games against strong bots within 30 days would be one possible condition; How? The clicking fastest to accept a game match is still the problem. -- robert jasiek ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
On Tue, Jan 3, 2012 at 2:27 PM, Robert Jasiek jas...@snafu.de wrote: On 03.01.2012 13:58, Ingo Althöfer wrote: 10 games against strong bots within 30 days would be one possible condition; How? The clicking fastest to accept a game match is still the problem. So the problem is that too many humans want to play strong bots? Perhaps it would help if kgs provided an option in the kgsgtp config to limit the rank difference where challenges are accepted. Humans can be picky about their opponents; maybe bots should also be allowed into that game :-) Erik ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
On 03/01/2012 13:55, Erik van der Werf wrote: On Tue, Jan 3, 2012 at 2:27 PM, Robert Jasiekjas...@snafu.de wrote: On 03.01.2012 13:58, Ingo Althöfer wrote: 10 games against strong bots within 30 days would be one possible condition; How? The clicking fastest to accept a game match is still the problem. So the problem is that too many humans want to play strong bots? Yes. Getting a game on KGS against a strong bot is difficult. It requires fast reactions to accept the game offer before any other user. We are considering ways of restricting the number of human entrants to a bot tournament on KGS. If such a tournament is ever held, it is likely that I will be running it. Rather than using rules such as 10 games against strong bots within 30 days, I will prefer: to use my discretion, giving preference to players such as BotHater, who I know to understand what is involved to seek applications only via this list to deter applications, by pointing out that eight solid hours of play against opponents who do not make yose mistakes will not be fun. My objective will be to avoid people who think that playing in a bot tournament will be cool, and then quit when they find that it isn't. Ordinary human KGS tournaments have many quitters - see e.g. https://www.gokgs.com/tournEntrants.jsp?sort=sid=625 Perhaps it would help if kgs provided an option in the kgsgtp config to limit the rank difference where challenges are accepted. Humans can be picky about their opponents; maybe bots should also be allowed into that game :-) There are many options which could usefully be added to the KGS bot-server interface. You are certainly not the first person to have suggested this one. I think it unlikely that any of them will be implemented, at least within the next year. Nick -- Nick Wedd n...@maproom.co.uk ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
On 03.01.2012 15:37, Petr Baudis wrote: My hope is that the games might finally become beautiful to watch They are. -- robert jasiek ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
As a human player, I would be opposed to playing against a program in a tournament. So I don't wish to participate in human tournaments with Crazy Stone. Also, having to operate the program manually would be a huge pain. Playing casual games automatically on KGS is much more pleasant for me, and not less interesting to watch than tournament games. Regardless of time control, the level of play is immensely higher than mine, anyway. Rémi On 3 janv. 2012, at 13:09, Ingo Althöfer wrote: Hello, I am in charge of organizing the computer go parts of the European Go Congress 2012. Before asking the main congress organizers if they were willing to allow a computer player in the main tournament I would like to make sure that - when allowed - really a programmer with a strong bot shows up. So, programmers of strong bots: Feel free to contact me soon (best by email - I will treat them confidential) if you are willing to come to Bonn. Your machine may be remote, but you or a member of your team should be in Bonn to operate the program. Some of the strong programs that come to my mind as possible participants are (in rather random order) Zen, CrazyStone, Steenvreter (Bonn is not very far from the NL), Erica, ManyFaces, Pachi, MoGo, Aya, Gomorra. Of course, such a participation would mean that each human participant has to declare at the start of the tournament whether he/she is willing to be paired against the bot - and these declarations have to be obeyed. Also, computer participants should be excluded from prize money. Jouni wrote: ... I would love to see strong gobot playing in EGC 2012 main tournament. Of course there is lots of organizing thing to do, but those two evil gobots are good enough and still not yet too good to participate into serious human tournaments with long thinking times. EGC 2012 takes place in Bonn (Germany, Rhine area), starting on July 21, and ending on August 04, 2012. The open tournament has 10 rounds, see at: http://www.egc2012.eu/congress/tournaments/open-european-championship In case of a/one computer participant in the main tournament I am willing to help by providing a present for each human player who plays the computer. Ingo. -- Empfehlen Sie GMX DSL Ihren Freunden und Bekannten und wir belohnen Sie mit bis zu 50,- Euro! https://freundschaftswerbung.gmx.de ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
On Tue, Jan 3, 2012 at 10:39 AM, Rémi Coulom remi.cou...@free.fr wrote: As a human player, I would be opposed to playing against a program in a tournament. So I don't wish to participate in human tournaments with Crazy Stone. When I was bringing my chess program to human tournaments, I was in a sort of conflict of interest because I sympathized with the human players. I wanted to get some experience with human play because when developing I always played other programs or self-play but I felt bad when one of the players felt that they were manipulated into playing the computer. Many did NOT sign the no-computer list simply because they forgot or were not asked to or knew the odds were low that they would play. My worst experience was in one of the US opens where my program was allowed to play only in the main tournament, but most of the side events including the speed chess tournaments I was not allowed to play. I payed a lot of money to travel that year and to get a lot of games with a lot of feedback from humans and felt cheated. I'm afraid that human and computers don't mix. Or at least they don't mix unless someone is willing to play them. Also, having to operate the program manually would be a huge pain. Playing casual games automatically on KGS is much more pleasant for me, and not less interesting to watch than tournament games. Regardless of time control, the level of play is immensely higher than mine, anyway. Rémi On 3 janv. 2012, at 13:09, Ingo Althöfer wrote: Hello, I am in charge of organizing the computer go parts of the European Go Congress 2012. Before asking the main congress organizers if they were willing to allow a computer player in the main tournament I would like to make sure that - when allowed - really a programmer with a strong bot shows up. So, programmers of strong bots: Feel free to contact me soon (best by email - I will treat them confidential) if you are willing to come to Bonn. Your machine may be remote, but you or a member of your team should be in Bonn to operate the program. Some of the strong programs that come to my mind as possible participants are (in rather random order) Zen, CrazyStone, Steenvreter (Bonn is not very far from the NL), Erica, ManyFaces, Pachi, MoGo, Aya, Gomorra. Of course, such a participation would mean that each human participant has to declare at the start of the tournament whether he/she is willing to be paired against the bot - and these declarations have to be obeyed. Also, computer participants should be excluded from prize money. Jouni wrote: ... I would love to see strong gobot playing in EGC 2012 main tournament. Of course there is lots of organizing thing to do, but those two evil gobots are good enough and still not yet too good to participate into serious human tournaments with long thinking times. EGC 2012 takes place in Bonn (Germany, Rhine area), starting on July 21, and ending on August 04, 2012. The open tournament has 10 rounds, see at: http://www.egc2012.eu/congress/tournaments/open-european-championship In case of a/one computer participant in the main tournament I am willing to help by providing a present for each human player who plays the computer. Ingo. -- Empfehlen Sie GMX DSL Ihren Freunden und Bekannten und wir belohnen Sie mit bis zu 50,- Euro! https://freundschaftswerbung.gmx.de ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
On 03/01/2012 15:19, terry mcintyre wrote: Does KGS support the ability to challenge a specific opponent? Yes. Could computer programs respond to such challenges, modulo specific criteria such as player rank, games per period, notoriety, etc? Bots playing on KGS cannot, I believe, accept existing challenges. Nick Terry McIntyre terrymcint...@yahoo.com Unix/Linux Systems Administration ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go -- Nick Wedd n...@maproom.co.uk ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
On Tue, Jan 03, 2012 at 04:02:18PM +, Nick Wedd wrote: On 03/01/2012 15:19, terry mcintyre wrote: Does KGS support the ability to challenge a specific opponent? Yes. Could computer programs respond to such challenges, modulo specific criteria such as player rank, games per period, notoriety, etc? Bots playing on KGS cannot, I believe, accept existing challenges. They can, using mode=wait and the 'opponent' variable. With little work, it would be possible to make e.g. a web-based challenging system where you could choose challengers in a more controlled way than KGS offers. kgsGtp would terminate after each game and be restarted with a newly generated config file with the appropriate opponent settings. It's just a matter of spending time on it. I didn't feel the incentive since Pachi is not so strong yet. ;-) -- Petr Pasky Baudis The goal of Computer Science is to build something that will last at least until we've finished building it. ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
The can take place in the auto pairing and choose some of the criteria as far as I know. Am Dienstag, den 03.01.2012, 16:02 + schrieb Nick Wedd: On 03/01/2012 15:19, terry mcintyre wrote: Does KGS support the ability to challenge a specific opponent? Yes. Could computer programs respond to such challenges, modulo specific criteria such as player rank, games per period, notoriety, etc? Bots playing on KGS cannot, I believe, accept existing challenges. Nick Terry McIntyre terrymcint...@yahoo.com Unix/Linux Systems Administration ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
That would be great idea. it would be interesting to watch. Although with slow thinking times, I feel sad for the gobots, because they do not have a chance against any of those. Anyway, this kind of matches would be great to watch, because there might be surprises, still. –Jouni On 3 January 2012 15:11, Jacques Basaldúa jacq...@dybot.com wrote: Maybe a tournament is not the best way to see quality computer/human games. There are better ways to measure the computer/computer performance and the human/human performance is not interesting here. We could simply schedule computer/human games on KGS (e.g., 3 times a year, one in each time zone afternoon) with around 4 KGS 5d+ humans and the 2 bots Zen and CrazyStone. Obvious human candidates are: Aja Huang, Robert Jasiek, Stefan Kaitschick, BotHater (don't know his name). Humans could use this list to subscribe and the pairings could be listed in advance. I don't think there are masses of KGS 5d+ players. It would be fun to watch. ** ** Jacques. ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] win rate bias and CLOP
Yes, I understood this. The particular case I was trying to optimize used two tunables, and it looked like there were two optimal combinations, but CLOP chose a point between them that was less optimal. I was checking the interaction between the beta constant and the MFGO bias. -Original Message- From: computer-go-boun...@dvandva.org [mailto:computer-go- boun...@dvandva.org] On Behalf Of Rémi Coulom Sent: Tuesday, January 03, 2012 7:19 AM To: computer-go@dvandva.org Subject: [Computer-go] win rate bias and CLOP It is important to understand that CLOP claims very little in terms of win rate. That is to say the win rate estimates it reports are all biased. Win rate over all samples underestimates the real win rate. Win rate near the maximum (central, and weighted) tend to be over-estimated. CLOP finds the location in parameter space that has the highest win rate. It may be the highest because it is the best, but also because it is the most lucky. That's why it is necessarily biased toward optimistic values. If the win rate over all samples is an improvement, then you can be sure you have an improvement. Otherwise you cannot be sure unless you actually play a lot of games with the suggested parameters. Rémi On 3 janv. 2012, at 14:09, Ingo Althöfer wrote: Hi David, David Fotland on CLOP-optimization: I tried it, but got no benefit so far. It claimed to find better settings for most parameters, but when I used them the program wasnt any stronger. Interestant. Had it similar strength or did it even become weaker? How often did the move proposals by your older ManyFaces and the CLOP-MF differ? Ingo. -- NEU: FreePhone - 0ct/min Handyspartarif mit Geld-zurück-Garantie! Jetzt informieren: http://www.gmx.net/de/go/freephone ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
[Computer-go] cgos client
I'm having a problem running the client for non 9x9. ./cgosview-linux-x86_32 -server cgos.boardspace.net -port 6813 could not execute Running it without any cli arguments works fine for 9x9 but trying to manually specify causes the error. Any tips? Ubuntu 32bit -Josh ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
Re: [Computer-go] win rate bias and CLOP
I am very enthusiastic about CLOP tuning. I overcame some roadblocks along the way that I want to share with you. You want CLOP to optimize strength, but it is actually optimizing Strength + Luck + Avoidance + Exploitation. Using CLOP effectively requires mitigating the last three factors. BTW, I imagine that CLOP could be any fully automated parameter tuning solution. That is, nothing here is really specific to CLOP. It just happens that CLOP is the first fully automated parameter tuning system that I have made to work. LUCK Remi has diagnosed the Luck factor: the win rate of the optimal setting is probably overestimated. This is not a big deal, provided that tuning runs go long enough for average and optimum win rates to be close together. (E.g., a few rating points.) If you change parameters to the ones recommended by CLOP, then the next CLOP run might claim that your program is weaker than before. This is just the luck vanishing, so I am not tempted to revert parameter settings in such cases. A more subtle point is that that CLOP does not measure the parameter combination that it recommends. The recommendation is a weighted average of points that it has measured, which converges (we think) to the optimum when the win-rate is a smooth function of the parameters. Actual performance can vary from projected, especially if the win-rate is not a smooth function. For example, Pebbles tunes against 2 opponents using 105 starting positions that are played with each color, making 420 initial situations. Both opponents and Pebbles are pseudo-random, so there is a great deal of variety from each initial situation. Still, you can imagine that the win rate is not entirely a smooth function of parameters. Maybe if you change a parameter by a small amount, then 10 similar initial situations will switch from wins to losses. That is just a reflection of how tuning is performed, so I just accept the recommended changes. Doing anything else would drive me crazy. Instead, I just accept that additional tuning will probably improve results. Basically: do sufficiently long runs (maybe 20K to 30K games when tuning 2 or 3 parameters?) until Average Optimal Average + K rating points (maybe K = 5?). And then blindly accept the new parameters without worrying about it. AVOIDANCE - Avoidance means: your program has weaknesses, and CLOP can tune parameters so that weaknesses are less likely to trigger. Avoidance makes your program stronger to some extent. That is, by avoiding weaknesses you can play better. But there are obvious limitations, as you cannot expect opponents to cooperate, but your search engine will use your own play to model the opponent. Pebbles has about 60 parameters, and I tuned them in groups of 2 or 3 parameters for about a dozen runs. Pebbles win rate rose from 47% to 55%. I was supremely happy, because that would be a good year and CLOP did it in just a month. But then I integrated some bug fixes, and the win rate dropped to 48%. What went wrong? For several weeks I was convinced that I had broken something, but I was unable to find anything. I verified every change using diffs, and I restored parameters, but was unable to make the win rate return to 56%. What I think happened is that tuning had tweaked Pebbles out to such an extent that it was now very sensitive to perturbations. There are 420 starting situations and ~60 parameters, so tuning each parameter just has to switch a few wins to push the win rate really high. I had fixed a few bugs that I discovered while the tuning was going on, and that was enough of a perturbation to make the win rate plummet. Figuring this out took a long time, so I now have a don't make yourself crazy policy here, too. I fix bugs as I find them, and integrate bug fixes into the tuning version ASAP. Now if the win rate drops suddenly, then there is an excellent chance that my last change was incorrect. But it hasn't been very long, so that's easy to find. BTW, tuning makes bugs easier to find. Your program is usually operating with carefully selected parameters. The tuning process creates an altogether different distribution of positions, which tends to expose logical errors. The result is that I have an endless supply of bugs to fix. EXPLOITATION Exploitation: tuning will create situations that the opposition handles badly. There is the same potential for good and bad as Avoidance, but the potential is reduced by having multiple training opponents. Pebbles trains against two opponents, and I will add others as Windows builds become available. Is it true that Pachi was tuned against Fuego, which was tuned against Mogo, which was tuned against GnuGo? If so, then game theory suggests that tuning against all of them will make Pebbles less susceptible to Exploitation and Avoidance defects. Using some self-play games in tuning should also help reduce both Avoidance and Exploitation. I have not tested that because using self-play
Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen
It was weaker, but I have to do more experiments. I didn’t compare move proposals. I just ran 2000-game tournaments vs gnugo before and after. Win rage went from 84.9% to 78.9%. I need to try it again with fewer variables and more games. I still think it is a great tool. David -Original Message- From: computer-go-boun...@dvandva.org [mailto:computer-go- boun...@dvandva.org] On Behalf Of Ingo Althöfer Sent: Tuesday, January 03, 2012 5:09 AM To: computer-go@dvandva.org Subject: Re: [Computer-go] CrazyStone in the 5-dan footsteps of Zen Hi David, David Fotland on CLOP-optimization: I tried it, but got no benefit so far. It claimed to find better settings for most parameters, but when I used them the program wasn�t any stronger. Interestant. Had it similar strength or did it even become weaker? How often did the move proposals by your older ManyFaces and the CLOP-MF differ? Ingo. -- NEU: FreePhone - 0ct/min Handyspartarif mit Geld-zur�ck-Garantie! Jetzt informieren: http://www.gmx.net/de/go/freephone ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go ___ Computer-go mailing list Computer-go@dvandva.org http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
