Re: [computer-go] creating a "random" position
On 7/8/07, Paul Pogonyshev <[EMAIL PROTECTED]> wrote: George Dahl wrote: > How would one go about creating a random board position with a uniform > distribution over all legal positions? Is this even possible? I am > not quite sure what I mean by uniform. If one flipped a three sided > coin to determine if each vertex was white,black or empty, then one > would have to deal with stones with no liberties somehow. Could those > just removed? As I remember from theory of probability, you can create such a uniformly "random" position this way[1]: 1. create a really random position, i.e. traverse all intersection and assign a black/white/empty state at random to each; 2. if it happens to be not legal, discard and repeat step 1. I believe it should be very fast, and this mustn't be difficult to check. I.e. rate of discards should be low enough for speed of algorithm to be speed of step 1 times C, where C is small. However, this will tend to give you very artificial-looking positions. Whether it is fine for your use-case, you know better. [1] http://en.wikipedia.org/wiki/Rejection_sampling Paul I think this is what I want. Thanks! So I might have to repeat this a few hundred times to actually get a legal position? - George ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
RE: [computer-go] No money in computer go?
But in fact it is a completely amateuristic field > where even the most basic > things are missing. As a chess programmer I did not even > think about, that > it is a problem to get a good game collection. 30,000 or more professional 19x19 games are readily available from SmartGo, MasterGo, or GoGOD. 9x9 games are not available because no one plays 9x9 :) There are no proper > interfaces, no serious tournaments, You came too late for the serious tournaments. Those started in 1986, and pretty much stopped in 2002 after the North Koreans were found cheating. Over $200,000 in prize money was given away. For example: http://www.reiss.demon.co.uk/webgo/fost99.htm First prize was 1 M yen, about $8,000. a wired data standard... > AND there is no money involved: I'd estimate the total go income to the older established go programs is at least $5 million, and probably closer to $10 million (over a period of 15 years or so). So there is plenty of money available. > > If it would be really a big challenge, there would be some > money. You missed the big challenge (over $1 million for a professional level program, with smaller prizes for milestones along the way). It was terminated in 2000 after Mr. Ing, the sponsor, died. > > Chrilly -David ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
RE: [computer-go] 9x9 games wanted and the next big challenge
However GTP was way better than what > preceded it and yet even the top programmers believed GMP was > sent by god and anything else was blasphemy. I have to object to this characterization :) GMP was very good at what is was designed to do, which was to allow people to play using a 1200 baud modem, before the internet was invented. This we very useful, since there were many people in the US that had no local club. Because the link was slow and unreliable, GMP was binary, and had reties, and provision for sending text messages, etc. Once the internet was invented, GMP was immediately obsolete, but because so many programs had implemented it, it became a de-facto standard for computer go tournaments. I think tournament organizers continued to specify it since it was too hard for organizers to invent something different. > > Even things like time-control systems are very logical in > Chess, but not in Go. Time control in go is quite logical if you remember that it is traditional and was invented before electronics. You can't do fisher time control by hand. I think go puts more emphasis on tradition than go, so things that were very logical and practical before computers, are still being used. Even though computers and electronic timers enable better approaches. The traditional ranking system of Go > isn't very rational > although it's understandable how it evolved. The ranking system is also very logical if you remember that it is intended to be maintained by hand, without electronic assistance. It's very easy to track the handicap I use with the people I play with most often, and change the handicap after a few consecutive wins or losses. Chess doesn't have a similar handicap system, so it has to rank based on probability of winning. Go, instead, adjusts that handicap until the winning probability is 50%. > It may be that because GO is more of a right brained activity, it > appeals more to the emotional, visual type of person. These kind of > people are probably a bit more into the culture and history of a game > than in the pure mathematical game itself. (There are also chess > players who love the culture of chess more than the game itself.) I think you are right. I don't see many chess players talking about creating a beautiful game. > > Also, Chess has evolved more recently, there have been > fundamental rule > changes within the last 2 or 3 hundred years I believe. Go also has many recent rule changes. Go was traditionally played without written rules. Codifying the traditions is what makes Japanese rules so complex. Take a look at the Ing Ko rule sometime :) David ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] creating a "random" position
On Jul 8, 2007, at 2:38 PM, Paul Pogonyshev wrote: George Dahl wrote: How would one go about creating a random board position with a uniform distribution over all legal positions? Is this even possible? I am not quite sure what I mean by uniform. If one flipped a three sided coin to determine if each vertex was white,black or empty, then one would have to deal with stones with no liberties somehow. Could those just removed? As I remember from theory of probability, you can create such a uniformly "random" position this way[1]: 1. create a really random position, i.e. traverse all intersection and assign a black/white/empty state at random to each; 2. if it happens to be not legal, discard and repeat step 1. I believe it should be very fast, and this mustn't be difficult to check. The check is easy: play all the stones (in any order, e.g., whatever order you have the points indexed) and see if there are any captures. If so, the position isn't legal. Peter Drake http://www.lclark.edu/~drake/ ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] creating a "random" position
i'd suggest that you need to consider whether what you really mean is a "position chosen from the uniform distribution of all legal go positions", or if you mean a position from somewhere near the middle game. (i.e. would you be comfortable with a board with 4 stones on it as one of these uniformly chosen boards?). do a single playout from an empty board with a normal distribution centered around whatever you think that the "average number of moves" is for a game to decide when to stop playing out)[1]. this isn't a uniformly chosen board position, but it will very likely accomplish whatever it is that you want to accomplish. s. [1] alternatively, you could have a fixed probability that you "stop playing out" that you check/evaluate after every new move. Yahoo! oneSearch: Finally, mobile search that gives answers, not web links. http://mobile.yahoo.com/mobileweb/onesearch?refer=1ONXIC ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] 9x9 games wanted and the next big challenge
steve uurtamo said: >> I have read dozens of times that computer-Go is the next big >> challenge. But in fact it is a completly amateuristic field where even >> the most basic things are missing. > > one thing that it seems to have plenty of is chess programmers who are > shocked and surprised that their lack of knowledge about the game of go > somehow seems to affect their ability to write a go playing program. One chess player I taught (who quickly surpassed me) found many strategic principles common to both games. But the basic strategic principles of go are highly counterintuitive to most chess players. (I like to think that much of the tone of oriental culture may have originated in efforts to make sense of go strategy...) There is such a thing in chess as "a premature attack," but not to the extent where leaving an opponent alone (not to "strengthen" him by attacking) in an area you hoped to attack later would be a common consideration. A tempo can be important in a chess game, but it's seldom as important as a piece--and sacrifices are much more natural in a game like go, where a tempo and a piece are the very same thing. I see strong players using the center a lot in modern go games, but the center in chess is critical while in go it's just another area. Sente is crucial in go, but no one gets to keep it throughout a game--and pressing an opponent with no particular object in mind is more likely to backfire than pay off. For whatever reason, good chess players often have trouble catching on to good style, although many chess-playing skills prove useful later. One major difference is in how far ahead a player can (and sometimes must) read out a position--ladders being the main example. (Some 30 moves ahead without a branch, or with perhaps one or two--and the human advantage is the ability to automatically reject the many irrelevant branches a program would need to consider in such positions.) Forrest Curo - This email was sent using AIS WebMail. http://www.americanis.net/ ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] creating a "random" position
In normal board size 1% random positions is legal, so it needs some rounds, but method is still superior propably any other if position must be random. t. Harri - Original Message - From: "Paul Pogonyshev" <[EMAIL PROTECTED]> To: Sent: Monday, July 09, 2007 12:38 AM Subject: Re: [computer-go] creating a "random" position George Dahl wrote: How would one go about creating a random board position with a uniform distribution over all legal positions? Is this even possible? I am not quite sure what I mean by uniform. If one flipped a three sided coin to determine if each vertex was white,black or empty, then one would have to deal with stones with no liberties somehow. Could those just removed? As I remember from theory of probability, you can create such a uniformly "random" position this way[1]: 1. create a really random position, i.e. traverse all intersection and assign a black/white/empty state at random to each; 2. if it happens to be not legal, discard and repeat step 1. I believe it should be very fast, and this mustn't be difficult to check. I.e. rate of discards should be low enough for speed of algorithm to be speed of step 1 times C, where C is small. However, this will tend to give you very artificial-looking positions. Whether it is fine for your use-case, you know better. [1] http://en.wikipedia.org/wiki/Rejection_sampling Paul ___ 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/
Re: [computer-go] 9x9 games wanted and the next big challenge
On Sun, 2007-07-08 at 11:23 -0700, Brian Slesinsky wrote: > It seems to me that a domain where "everything is so amateuristic" has > its advantages, if you can only see them. Here is a field that is > small enough that most people know each other and anyone can > contribute with a certain amount of effort. These are the early days; > computer go's best years are surely yet to come. And yet it is not so > early that progress is slow and there is little hope. Isn't that > better than working in an area where everything has been done? > > I don't follow computer chess, but my naive outsider's perception is > that it is largely solved. Perhaps those who know more about it can > say more. Chess is far from being solved. There is still a LOT of room at the top when one program can still dominate another. The current best chess program is significantly stronger than the second best and I think it likely that a perfect chess player would dominate this best program. It would rarely even have to suffer a draw against Rybka. Lots of room for improvement still. What is true is that the very best chess program are now better than people. This used to be considered the "gold standard", the ultimate but of course we now know that was foolish. I went over to GO for several reasons. I felt that computer chess had becomes heavily dominated by engineers. There is still some room for imagination, but not as much. It's mostly knowledge engineering, programming tricks and fine tuning. You still must come to every tournament with the fastest possible computer, preferably with multi-processors. Whereas GO is something altogether different. A UCT type breakthrough is probably not possible in Chess. Maybe some interesting good ideas are still left, but nothing dramatic.I left chess and leave it for the engineers to do their thing. The Go community is also far better behaved. A fight can still break out, but it's nothing like what happens in Chess, where there are decades long grudges and bitter wars of words. You cannot have an un-moderated chess group. Chrilly is certainly right about how Chess programmers perceive things. Although there is conservatism in every thing including computer chess, the Go community as a whole is rather old fashion and tends to shoot themselves in the foot when it comes to any kind of progress. Almost like old men who think the way their grand-daddy did it is good enough for them.Most of us probably remember the heavy resistance to GTP, which in itself is inferior to UCI, the universal chess interface, which is the GO equivalent of GTP. However GTP was way better than what preceded it and yet even the top programmers believed GMP was sent by god and anything else was blasphemy. One surprise is that SGF was accepted by the go community. In some ways SGF is technically superior to PGN.However PGN is actually far more practical. The computer Chess community seems to value anything that is practical, the computer Go community seems to embrace anything that isn't. I think that's what Chrilly has noticed. Even things like time-control systems are very logical in Chess, but not in Go. The traditional ranking system of Go isn't very rational although it's understandable how it evolved. But the Chess community is usually very quick to discard the old if something more practical comes along. The Go community is far slower at embracing chess that is good. It may be that because GO is more of a right brained activity, it appeals more to the emotional, visual type of person. These kind of people are probably a bit more into the culture and history of a game than in the pure mathematical game itself. (There are also chess players who love the culture of chess more than the game itself.) Also, Chess has evolved more recently, there have been fundamental rule changes within the last 2 or 3 hundred years I believe. Somehow this has translated to a more conservative view of even computer chess and how it should be done. - Don > - Brian > ___ > 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/
Re: [computer-go] creating a "random" position
George Dahl wrote: > How would one go about creating a random board position with a uniform > distribution over all legal positions? Is this even possible? I am > not quite sure what I mean by uniform. If one flipped a three sided > coin to determine if each vertex was white,black or empty, then one > would have to deal with stones with no liberties somehow. Could those > just removed? As I remember from theory of probability, you can create such a uniformly "random" position this way[1]: 1. create a really random position, i.e. traverse all intersection and assign a black/white/empty state at random to each; 2. if it happens to be not legal, discard and repeat step 1. I believe it should be very fast, and this mustn't be difficult to check. I.e. rate of discards should be low enough for speed of algorithm to be speed of step 1 times C, where C is small. However, this will tend to give you very artificial-looking positions. Whether it is fine for your use-case, you know better. [1] http://en.wikipedia.org/wiki/Rejection_sampling Paul ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] creating a "random" position
At 21:54 08/07/2007, you wrote: I don't have such algorithm, you can count legal positions like: http://www.lysator.liu.se/~gunnar/legal.pike.txt Modifying it could provide some way select random position atleast for small boards. Ported that for java but not studied much of it yet, intresting anyway. t. Harri This page seems more up to date, and links a paper http://homepages.cwi.nl/~tromp/go/legal.html ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] creating a "random" position
Start with an empty board and do a large number (eg 1, depending on how accurate you need the uniformity) of 'operations' on it, where an operation is the following: 1) choose an intersection at random 2) toss a fair coin 3) change the occupation of the intersection according to the following table Old occupation Coin New occupation BlackH White WhiteH Empty EmptyH Black BlackT Empty WhiteT Black EmptyT White 4) If the new position is illegal, undo the change and go to 1) 5) Exit At 21:22 08/07/2007, you wrote: How would one go about creating a random board position with a uniform distribution over all legal positions? Is this even possible? I am not quite sure what I mean by uniform. If one flipped a three sided coin to determine if each vertex was white,black or empty, then one would have to deal with stones with no liberties somehow. Could those just removed? - George ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/ -- This email has been verified as Virus free Virus Protection and more available at http://www.plus.net ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] creating a "random" position
I don't have such algorithm, you can count legal positions like: http://www.lysator.liu.se/~gunnar/legal.pike.txt Modifying it could provide some way select random position atleast for small boards. Ported that for java but not studied much of it yet, intresting anyway. t. Harri - Original Message - From: "George Dahl" <[EMAIL PROTECTED]> To: "computer-go" Sent: Sunday, July 08, 2007 11:22 PM Subject: [computer-go] creating a "random" position How would one go about creating a random board position with a uniform distribution over all legal positions? Is this even possible? I am not quite sure what I mean by uniform. If one flipped a three sided coin to determine if each vertex was white,black or empty, then one would have to deal with stones with no liberties somehow. Could those just removed? - George ___ 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/
Re: [computer-go] creating a "random" position
On 7/8/07, George Dahl <[EMAIL PROTECTED]> wrote: How would one go about creating a random board position with a uniform distribution over all legal positions? Is this even possible? I am not quite sure what I mean by uniform. If one flipped a three sided coin to determine if each vertex was white,black or empty, then one would have to deal with stones with no liberties somehow. Could those just removed? - George All legal positions can be enumerated, so just create a database containing all legal positions and then select one at random. If this does not work for you, e.g., due to insufficient storage, just keep generating random positions (using your special coin) until you hit one that's legal. OC, once you start 'correcting' illegal positions it probably won't be uniform over all legal positions any more, that's why, unless you come up with something clever, you should regenerate the entire position. May I ask why you need any of this? E. ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] 9x9 games wanted and the next big challenge
It seems to me that a domain where "everything is so amateuristic" has its advantages, if you can only see them. Here is a field that is small enough that most people know each other and anyone can contribute with a certain amount of effort. These are the early days; computer go's best years are surely yet to come. And yet it is not so early that progress is slow and there is little hope. Isn't that better than working in an area where everything has been done? Yes. The original meaning of amateur is lover. E.g. I enjoyed the athmosphere when I was operating GoAhead in the olympiad 2003. Its also known that humans generally evaluate/feel the difference and not the absolute level. So its nicer to be in the non-saturated point. But as professional its a job and one can not completly ignore mundane tasks like the Euro/h. The formula: There is money for everything what is important, and if there is no money, it is not important, is certainly also wrong. A counter-example is the research for Leprosy-medicaments. They ones who have Leprosy have no money and there is no incentive for the pharma-companies to invest. But also academic institutions do almost no research. There are no funds from industry. I don't follow computer chess, but my naive outsider's perception is that it is largely solved. Perhaps those who know more about it can say more. Its not solved in the theoretical sense. God could certainly give them 2 pawns as handicap. But it is solved from the practical sense, because God could give the top-humans a knight ahead. The only way to measure the difference between Rybka and Fritz is to let them play against each other. Just looking on the play of each of them or playing against them, most humans would not be able to say: Rybka is 100 Elo stronger. Even Topalov does not play nowadays for fun in the evening some blitz-games against a programm. Although he likes challenges, he neither runs with his head against the wall in his living room to check who is stronger. Most of the top-GMs hate the programms, because the size of opening theory has become a nightmare. Some opening lines are practically fully analysed and hence not playable anymore. I know some top players who would like to ban computers for preperation. But its impossible to check such a ban. Chrilly . , so also these GMs use very heavily PCs. - Brian ___ 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/
[computer-go] creating a "random" position
How would one go about creating a random board position with a uniform distribution over all legal positions? Is this even possible? I am not quite sure what I mean by uniform. If one flipped a three sided coin to determine if each vertex was white,black or empty, then one would have to deal with stones with no liberties somehow. Could those just removed? - George ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] 9x9 games wanted and the next big challenge
David, Very well said. Thank you. Jim David Doshay wrote: Chrilly, It is hard to disagree with what Jim writes, but I will in a small way. When I recently flew to Asia, the screen on the seatback in front of me offered Go as one of its games. At its highest level it played far worse than the average program on CGOS or in a KGS computer tournament, and yet somebody was paid by that airline for the use of that program. And Go++ makes a living for its programmer. There is money to be made in computer Go, but as Jim states, right now the risk/reward ratio does not encourage most normal investors, so it is for either 1) those with a high risk threshold, 2) those who think more about research than production, 3) those motivated by how hard it is and not put off by how much effort it is going to take, or 4) programmers of other games who underestimate how hard it really is. Please do not take offense by number 4. I have huge respect for your programming ability and am glad that you have joined us. Cheers, David On 8, Jul 2007, at 8:36 AM, Jim O'Flaherty, Jr. wrote: Chrilly, The purpose of investment is to generate a return exceeding the original investment, i.e. a profit. Given the state of Go, I am finding it difficult to imagine why an investor would choose to put any good money into Go. There is absolutely no reliable expectation that Go will achieve even close to strong amateur status (1D) in the next couple of years. It's possible some wealthy person might decide to generously donate money into the computer Go domain so as to forward his own passion, just as many of the people here generously donate their own very valuable personal time. Go is not a reasonable place to put investments. At present and from everything I can see, computer Go development depends upon personal passion and generosity. And sans a huge breakthrough, I am currently unable to see this changing anytime soon. That said, I think once Go AI becomes sufficiently and robustly skilled to reliably start giving strong amateurs (>1D) genuinely competitive games, you will start to see investment rise. And given a sufficiently high enough rate of change (objectively measured as increases in playing skill), you will start to see the investments accelerate as competition will spur on more innovation resulting in more successes resulting in more investment resulting in further innovation...and a positive feedback loop will be boot strapped. As the probability of producing profits rise, the risk around insufficient returns on an investment fall. Eventually a threshold is crossed and the system becomes self-generative. Succinctly put - there is no money in computer Go (at least compared to computer Chess) because there is currently no hope (mathematically speaking) of the existing crop of computer Go programs to scale up to anything less than moderate amateur levels. Once this changes from no hope to a remote possibility, the investment around Go will likely follow. No to be too "Zen" here, but...the sooner you accept things as they are and stop resisting "what is", the sooner you become free to move forward. Go investment is working exactly as it ought, in relation to the "whole". Finally, thank you for your contribution to computer Go. I get that it is an act of generosity (realistically, what else could it possibly be). And I personally appreciate it. Jim ___ 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/
Re: [computer-go] 9x9 games wanted and the next big challenge
Chrilly, It is hard to disagree with what Jim writes, but I will in a small way. When I recently flew to Asia, the screen on the seatback in front of me offered Go as one of its games. At its highest level it played far worse than the average program on CGOS or in a KGS computer tournament, and yet somebody was paid by that airline for the use of that program. And Go++ makes a living for its programmer. There is money to be made in computer Go, but as Jim states, right now the risk/reward ratio does not encourage most normal investors, so it is for either 1) those with a high risk threshold, 2) those who think more about research than production, 3) those motivated by how hard it is and not put off by how much effort it is going to take, or 4) programmers of other games who underestimate how hard it really is. Please do not take offense by number 4. I have huge respect for your programming ability and am glad that you have joined us. Cheers, David On 8, Jul 2007, at 8:36 AM, Jim O'Flaherty, Jr. wrote: Chrilly, The purpose of investment is to generate a return exceeding the original investment, i.e. a profit. Given the state of Go, I am finding it difficult to imagine why an investor would choose to put any good money into Go. There is absolutely no reliable expectation that Go will achieve even close to strong amateur status (1D) in the next couple of years. It's possible some wealthy person might decide to generously donate money into the computer Go domain so as to forward his own passion, just as many of the people here generously donate their own very valuable personal time. Go is not a reasonable place to put investments. At present and from everything I can see, computer Go development depends upon personal passion and generosity. And sans a huge breakthrough, I am currently unable to see this changing anytime soon. That said, I think once Go AI becomes sufficiently and robustly skilled to reliably start giving strong amateurs (>1D) genuinely competitive games, you will start to see investment rise. And given a sufficiently high enough rate of change (objectively measured as increases in playing skill), you will start to see the investments accelerate as competition will spur on more innovation resulting in more successes resulting in more investment resulting in further innovation...and a positive feedback loop will be boot strapped. As the probability of producing profits rise, the risk around insufficient returns on an investment fall. Eventually a threshold is crossed and the system becomes self-generative. Succinctly put - there is no money in computer Go (at least compared to computer Chess) because there is currently no hope (mathematically speaking) of the existing crop of computer Go programs to scale up to anything less than moderate amateur levels. Once this changes from no hope to a remote possibility, the investment around Go will likely follow. No to be too "Zen" here, but...the sooner you accept things as they are and stop resisting "what is", the sooner you become free to move forward. Go investment is working exactly as it ought, in relation to the "whole". Finally, thank you for your contribution to computer Go. I get that it is an act of generosity (realistically, what else could it possibly be). And I personally appreciate it. Jim ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] 9x9 games wanted and the next big challenge
Much of the work going on happens in universities and is funded just like other work there. It probably adds up to some real money when you consider all of the salaries and equipment. But there isn't always a line item called "computer go." All that research money that went into chess once upon a time wasn't because the world needed a better chess program as an end in itself. I wonder how the sponsors feel about their return on investment wrt computer chess. - Dave Hillis -Original Message- From: David Doshay <[EMAIL PROTECTED]> To: computer-go Sent: Sun, 8 Jul 2007 1:54 pm Subject: Re: [computer-go] 9x9 games wanted and the next big challenge On 8, Jul 2007, at 2:51 AM, chrilly wrote:? ? > If it would be really a big challenge, there would be some money.? ? According to Herodotus "The Histories" right after king Xerxes of Persia lost 20,000 men at Thermopylae fighting 300 Spartans and a collection of less than 100 others, a few Arcadian deserters informed Xerxes that the Greeks were celebrating the Olympic festival. When asked what the prize was for which the men contended, they mentioned the wreaths of olive leaves. Tritantaechmes, upon hearing that the prize was not a large sum of money, cried out "Good heavens, what kind of men have you brought us to fight against - men who compete with one another for no material reward, but only for honour!"? ? About 20 years ago I asked another physics student who was known as a good programmer to join me in an attempt to win the Ing prize. His answer was "If we are good enough to win the Ing then we can make far more than a million in less time just working in Silicon Valley." He was right.? ? If it were easy it would have been done by now. It is NOT easy, and thus it IS a really big challenge.? ? Cheers,? David? ___? computer-go mailing list? [EMAIL PROTECTED] http://www.computer-go.org/mailman/listinfo/computer-go/? Check Out the new free AIM(R) Mail -- Unlimited storage and industry-leading spam and email virus protection. ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] 9x9 games wanted and the next big challenge
It seems to me that a domain where "everything is so amateuristic" has its advantages, if you can only see them. Here is a field that is small enough that most people know each other and anyone can contribute with a certain amount of effort. These are the early days; computer go's best years are surely yet to come. And yet it is not so early that progress is slow and there is little hope. Isn't that better than working in an area where everything has been done? I don't follow computer chess, but my naive outsider's perception is that it is largely solved. Perhaps those who know more about it can say more. - Brian ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] 9x9 games wanted and the next big challenge
On 8, Jul 2007, at 2:51 AM, chrilly wrote: If it would be really a big challenge, there would be some money. According to Herodotus "The Histories" right after king Xerxes of Persia lost 20,000 men at Thermopylae fighting 300 Spartans and a collection of less than 100 others, a few Arcadian deserters informed Xerxes that the Greeks were celebrating the Olympic festival. When asked what the prize was for which the men contended, they mentioned the wreaths of olive leaves. Tritantaechmes, upon hearing that the prize was not a large sum of money, cried out "Good heavens, what kind of men have you brought us to fight against - men who compete with one another for no material reward, but only for honour!" About 20 years ago I asked another physics student who was known as a good programmer to join me in an attempt to win the Ing prize. His answer was "If we are good enough to win the Ing then we can make far more than a million in less time just working in Silicon Valley." He was right. If it were easy it would have been done by now. It is NOT easy, and thus it IS a really big challenge. Cheers, David ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] 9x9 games wanted and the next big challenge
Chrilly, The purpose of investment is to generate a return exceeding the original investment, i.e. a profit. Given the state of Go, I am finding it difficult to imagine why an investor would choose to put any good money into Go. There is absolutely no reliable expectation that Go will achieve even close to strong amateur status (1D) in the next couple of years. It's possible some wealthy person might decide to generously donate money into the computer Go domain so as to forward his own passion, just as many of the people here generously donate their own very valuable personal time. Go is not a reasonable place to put investments. At present and from everything I can see, computer Go development depends upon personal passion and generosity. And sans a huge breakthrough, I am currently unable to see this changing anytime soon. That said, I think once Go AI becomes sufficiently and robustly skilled to reliably start giving strong amateurs (>1D) genuinely competitive games, you will start to see investment rise. And given a sufficiently high enough rate of change (objectively measured as increases in playing skill), you will start to see the investments accelerate as competition will spur on more innovation resulting in more successes resulting in more investment resulting in further innovation...and a positive feedback loop will be boot strapped. As the probability of producing profits rise, the risk around insufficient returns on an investment fall. Eventually a threshold is crossed and the system becomes self-generative. Succinctly put - there is no money in computer Go (at least compared to computer Chess) because there is currently no hope (mathematically speaking) of the existing crop of computer Go programs to scale up to anything less than moderate amateur levels. Once this changes from no hope to a remote possibility, the investment around Go will likely follow. No to be too "Zen" here, but...the sooner you accept things as they are and stop resisting "what is", the sooner you become free to move forward. Go investment is working exactly as it ought, in relation to the "whole". Finally, thank you for your contribution to computer Go. I get that it is an act of generosity (realistically, what else could it possibly be). And I personally appreciate it. Jim chrilly wrote: Sil wrote: How about http://home.wwgo.jp/jp/minigo/ It seems that only 24 games are available. Is the whole collection available somewhere? Rémi I have read dozens of times that computer-Go is the next big challenge. But in fact it is a completly amateuristic field where even the most basic things are missing. As a chess programmer I did not even think about, that it is a problem to get a good game collection. There are no proper interfaces, no serious tournaments, a wired data standard... AND there is no money involved: For professional programming I get 60Euro/h (1Euro=1.35$). 2.000h x 60 = 120.000 Euro. This equation is of course completly wrong. One can not make in 2000h a very strong Go programm and one can not earn 120.000 Euro with it. A more realistic equation is; 20.000 Euro/5000h = 4Euro/h. The minimum wage (by law) is in Austria 6Euro/h. Obviously Go programming is even more unqualified than washing dishes in a restaurant. If it would be really a big challenge, there would be some money. In chess nowadays there is also no money. But once it was a good business and there was some considerable money for Deep Blue and on a smaller scale also for Hydra, there was Don's project at MIT, one got a big Cray for Cray-Blitz, Ken Thompson build a chess engine Its like some hobbyst engineers and hobby-pilots would try to fly to the moon. Its probably only good for to write some academic papers. In this case its even an advantage that everything is so amateuristic. The general level is low and one can be the one-eyed king under blind ones. Its clear to me that things are as they are in the West. Go is played only by a small freak community. But if it is so important in China/Korea/Japan why is'nt there something like Fritz and ChessBase? Or does it exist and we are living in a completly other Go-world? Chrilly P.S.: I do not want to offend anyone in this list. Everybody here does his best. I am just feed up with the things as they are. ___ 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/ ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] 9x9 games wanted and the next big challenge
3) seriousness can't be measured as the short term money you can make directly selling your work. I understand that you think that researchers are paid just to play writing useless papers for themself. But there are not more stupid than others, and maybe they think they are doing something useful, even if it can't be measured by the direct sell of what they produce. I think UCT is an major new idea. Like Alpha-Beta. I am not at all against scientific work or papers. I have myself written some of them and even succeeded to place one in the Journal of the American Statistical Association. And I also understand, that everybody has to present his work as very important. Otherwise other people who are better in this respect get the funds. And one can do only a good work, if one really believes in this. A classical quote from the German scientist Max Weber is: To be a good scientist one has to write every sentence as if the existence of the world depends on this sentence. (Although one of course knows that this is usually not the case). But science, the science world, is also a very closed world and there is a tendency for l'art pour l'art. I am just critizing this aspect. Chrilly ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] 9x9 games wanted and the next big challenge
> I have read dozens of times that computer-Go is the next big challenge. > But in fact it is a completly amateuristic field where even the most basic > things are missing. one thing that it seems to have plenty of is chess programmers who are shocked and surprised that their lack of knowledge about the game of go somehow seems to affect their ability to write a go playing program. it's as if they are so used to their success with chess that they simply cannot understand why and how it could come to pass that they aren't already writing the best go playing programs in the world. 9x9 is a learning game. it's as if you could teach children to play chess on a 5x5 board. no, there wouldn't be databases of thousands of professional such games, because professionals are mostly spending their time playing 19x19 go. the *only* reason that 9x9 go has such appeal in the computer go community is because it is more tractable. success is much more easily "measured" on a 9x9 board. but that's not the fault of the game, and is unrelated to how and on what size board most real-world games are played. > There are no proper interfaces, no serious tournaments, a wired data > standard... this is a strange claim. i wonder what you mean by "no proper interfaces" and "no serious tournaments". is a serious tournament one where a lot of money is at stake? or is it where the best programs in the world get together to compete? because the latter happens quite frequently. > AND there is no money involved: most people doing this aren't doing this to make money, although some are. as far as corporate support for a strong go-playing machine is concerned, or for very-well-funded tournaments (for instance), there hasn't been enough success on a 19x19 board for the *expected* result of such funding to result in games that anyone would want to watch. the go-playing community can pick up a daily newspaper and see a professional game of such quality that programs cannot hope to match at this point. why would the general public want to watch a public demonstration of a program that can be beaten by many 10-year-old children? and if, 6 months from now, someone has a 19x19 program that can beat an arbitrary 10-year-old child, will that be a spectacle worth paying to watch? > If it would be really a big challenge, there would be some money. the ~$1M prize stood for years without being touched. the development of a program/machine that could win that prize (i.e. beat an insei or a professional) would generate a tremendous about of interest in computer go, and i am quite sure that public challenges would be well-funded. keep in mind that in the professional go-playing community, tournament winnings can vary quite a bit -- only a few players could hope to make a living from one or two tournament wins -- everyone else must spend their free time teaching students for pay, writing articles, etc. maybe this isn't so different from the chess world after all. how strong were chess programs when the push for Deep Blue happened? were they the equivalent of ~1300 ELO weaker than professional play? s. The fish are biting. Get more visitors on your site using Yahoo! Search Marketing. http://searchmarketing.yahoo.com/arp/sponsoredsearch_v2.php ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] Re: 9x9 games wanted
Mr. Okasaki pointed out that the records may be the same as the records Chrilly already have. If so, I'm sorry. -gg Hideki Kato: <[EMAIL PROTECTED]>: >I've asked Japanese computer Go community if anyone have the database >(about 250 records of Professionals). > >Dr. Saito replied today. With Mr. Okasaki's acknowledgement for >research purposes, I will be able to have the records soon. Please >wait a few days. > >-gg > >Rémi Coulom: <[EMAIL PROTECTED]>: >>Sil wrote: >>> How about http://home.wwgo.jp/jp/minigo/ >>> >>It seems that only 24 games are available. Is the whole collection >>available somewhere? >> >>Rémi >>___ >>computer-go mailing list >>computer-go@computer-go.org >>http://www.computer-go.org/mailman/listinfo/computer-go/ >-- >[EMAIL PROTECTED] (Kato) >___ >computer-go mailing list >computer-go@computer-go.org >http://www.computer-go.org/mailman/listinfo/computer-go/ -- [EMAIL PROTECTED] (Kato) ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] Re: 9x9 games wanted
I've asked Japanese computer Go community if anyone have the database (about 250 records of Professionals). Dr. Saito replied today. With Mr. Okasaki's acknowledgement for research purposes, I will be able to have the records soon. Please wait a few days. -gg Rémi Coulom: <[EMAIL PROTECTED]>: >Sil wrote: >> How about http://home.wwgo.jp/jp/minigo/ >> >It seems that only 24 games are available. Is the whole collection >available somewhere? > >Rémi >___ >computer-go mailing list >computer-go@computer-go.org >http://www.computer-go.org/mailman/listinfo/computer-go/ -- [EMAIL PROTECTED] (Kato) ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] 9x9 games wanted and the next big challenge
In message <[EMAIL PROTECTED]>, alain Baeckeroot <[EMAIL PROTECTED]> writes Le dimanche 8 juillet 2007 11:51, chrilly a écrit : If it would be really a big challenge, there would be some money. There was a computer challenge with 1 million dollar prize during many years, for a program abble to beat one professional choosen by the sponsor. I don't know if it is still valid offer. The prize was not exactly US $1,000,000, but Taiwanese $40,000,000. It would currently be worth US $1,219,000. When it was on offer, it could be won be a program beating an insei (a trainee professional, about as strong as a European amateur 6-dan) in four of seven games. Judging from past events, this would not be a seven-game match between the program and an insei, but seven simultaneous games between the program and seven different inseis - this is rather easier for the program, as the inseis would have less scope to learn from its mistakes. There were smaller prizes, for the first program to beat an insei on various handicaps, as follows: Handicap match prize, NT$ no komi3/520,000,000 (about US $610,000) 2 stones 2/310,000,000 3 stones 2/3 5,000,000 4 stones 2/3 2,000,000 5 stones 2/3 1,000,000 6 stones 2/3 850,000 7 stones 2/3 700,000 8 stones 2/3 550,000 9 stones 2/3 400,000 10 stones 2/3 250,000 won by Handtalk in 1997 etc. These prizes are not currently on offer. They were last on offer in 2000. A report from the 2000 Ing Cup stated "There was a rumor that the Ing competition will be extended for 10 more years, but it was unconfirmed." This rumour must have been false. No Ing cup has been held since then. I have heard a rumour that the managers of the Ing Foundation may one day decide to reinstate the prize money; but as the years pass this is looking increasingly unlikely. Nick -- Nick Wedd[EMAIL PROTECTED] ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] Cgos problems and resignation
On Sun, 2007-07-08 at 12:53 +0200, Magnus Persson wrote: > I just had an exception in Valkyria because it recieved > > "play b resign" > > from the server. > > As far as I know CGOS used to to send nothing to the winner when a program > resigned. Am I wrong or has this something to do with the current stability > problems of CGOS? > > Her is the end of the log. Game was 69653 > > 12:35:43C->E genmove w > 12:35:45E->C = j1 > 12:35:45C->S j1 > 12:35:47S->C play b RESIGN 47144 > 12:35:47C->E play b RESIGN It's easy to take this out of the client. I think I have it in because RESIGN is a legitimate GTP move and part of the standard. And some programs log the games for themselves, there would be no way to log a complete games without this information - you would have an incomplete game record that just stops for no known reason. But I think this has been in there a long time - at least for the newer CGOS server. Let me think about this a bit. Perhaps I can add a command line switch to the client to prevent sending the last move if it is resign.Even though GTP says it's a "move" there might be other programs that don't understand it. - Don > > > ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] 9x9 games wanted and the next big challenge
Le dimanche 8 juillet 2007 11:51, chrilly a écrit : > If it would be really a big challenge, there would be some money. There was a computer challenge with 1 million dollar prize during many years, for a program abble to beat one professional choosen by the sponsor. I don't know if it is still valid offer. > In chess > nowadays there is also no money. But once it was a good business and there > was some considerable money for Deep Blue and on a smaller scale also for > Hydra, there was Don's project at MIT, one got a big Cray for Cray-Blitz, > Ken Thompson build a chess engine > Its like some hobbyst engineers and hobby-pilots would try to fly to the > moon. Titanic was build by professionals, and Noah's arch by an amateur ;-) (Kon Tiki is a more recent and scientific exemple of incredible amateurish success) > Its probably only good for to write some academic papers. In this case its > even an advantage that everything is so amateuristic. The general level is > low and one can be the one-eyed king under blind ones. If i remeber, last year you said something like "As a professional programmer, i don't want to ruin my reputation with a poor go program" :-) And the state of the art is: go programs are just dumb on 19x19, lots of research are needed, but more engeneering power would probably do nothing. > > Its clear to me that things are as they are in the West. Go is played only > by a small freak community. But if it is so important in China/Korea/Japan > why is'nt there something like Fritz and ChessBase? Or does it exist and we > are living in a completly other Go-world? Some dozens of 9x9 pro games are at http://gobase.org/9x9/ There are databases of nearly 5 pro games on 19X19, this should be good enough for some years in computer go. 9x9 is a teaching tool, or a fun tactical exercice, but it is not Go because of lack of strategy. Alain ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
[computer-go] Cgos problems and resignation
I just had an exception in Valkyria because it recieved "play b resign" from the server. As far as I know CGOS used to to send nothing to the winner when a program resigned. Am I wrong or has this something to do with the current stability problems of CGOS? Her is the end of the log. Game was 69653 12:35:43C->E genmove w 12:35:45E->C = j1 12:35:45C->S j1 12:35:47S->C play b RESIGN 47144 12:35:47C->E play b RESIGN -- Magnus Persson Berlin, Germany ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] 9x9 games wanted and the next big challenge
Hi Chrilly, 1) there are database of thousands of professional games for few dollards. There are not 9x9, but (i) making database is not making progress in the field, it is just having some temporary advantage in tournaments. (ii) Opening is much less important in Go than in Chess, it is why we are not so crazy about opening. At least at the current level of programs. (iii) 19x19 Go is the real game, and you can get as many games as you want, just clicking on three links. 2) interfaces are good enough for what we need, and tournaments as CGOS or KGS are good tools to take the current temperature of field. 3) seriousness can't be measured as the short term money you can make directly selling your work. I understand that you think that researchers are paid just to play writing useless papers for themself. But there are not more stupid than others, and maybe they think they are doing something useful, even if it can't be measured by the direct sell of what they produce. 4) I guess people that sell commercial programs are making money. All that said, I agree that computer go is certainly much less mature than computer chess. Sylvain I have read dozens of times that computer-Go is the next big challenge. But in fact it is a completly amateuristic field where even the most basic things are missing. As a chess programmer I did not even think about, that it is a problem to get a good game collection. There are no proper interfaces, no serious tournaments, a wired data standard... AND there is no money involved: For professional programming I get 60Euro/h (1Euro=1.35$). 2.000h x 60 = 120.000 Euro. This equation is of course completly wrong. One can not make in 2000h a very strong Go programm and one can not earn 120.000 Euro with it. A more realistic equation is; 20.000 Euro/5000h = 4Euro/h. The minimum wage (by law) is in Austria 6Euro/h. Obviously Go programming is even more unqualified than washing dishes in a restaurant. If it would be really a big challenge, there would be some money. In chess nowadays there is also no money. But once it was a good business and there was some considerable money for Deep Blue and on a smaller scale also for Hydra, there was Don's project at MIT, one got a big Cray for Cray-Blitz, Ken Thompson build a chess engine Its like some hobbyst engineers and hobby-pilots would try to fly to the moon. Its probably only good for to write some academic papers. In this case its even an advantage that everything is so amateuristic. The general level is low and one can be the one-eyed king under blind ones. Its clear to me that things are as they are in the West. Go is played only by a small freak community. But if it is so important in China/Korea/Japan why is'nt there something like Fritz and ChessBase? Or does it exist and we are living in a completly other Go-world? Chrilly P.S.: I do not want to offend anyone in this list. Everybody here does his best. I am just feed up with the things as they are. ___ 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/ ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
[computer-go] 9x9 games wanted and the next big challenge
Sil wrote: How about http://home.wwgo.jp/jp/minigo/ It seems that only 24 games are available. Is the whole collection available somewhere? Rémi I have read dozens of times that computer-Go is the next big challenge. But in fact it is a completly amateuristic field where even the most basic things are missing. As a chess programmer I did not even think about, that it is a problem to get a good game collection. There are no proper interfaces, no serious tournaments, a wired data standard... AND there is no money involved: For professional programming I get 60Euro/h (1Euro=1.35$). 2.000h x 60 = 120.000 Euro. This equation is of course completly wrong. One can not make in 2000h a very strong Go programm and one can not earn 120.000 Euro with it. A more realistic equation is; 20.000 Euro/5000h = 4Euro/h. The minimum wage (by law) is in Austria 6Euro/h. Obviously Go programming is even more unqualified than washing dishes in a restaurant. If it would be really a big challenge, there would be some money. In chess nowadays there is also no money. But once it was a good business and there was some considerable money for Deep Blue and on a smaller scale also for Hydra, there was Don's project at MIT, one got a big Cray for Cray-Blitz, Ken Thompson build a chess engine Its like some hobbyst engineers and hobby-pilots would try to fly to the moon. Its probably only good for to write some academic papers. In this case its even an advantage that everything is so amateuristic. The general level is low and one can be the one-eyed king under blind ones. Its clear to me that things are as they are in the West. Go is played only by a small freak community. But if it is so important in China/Korea/Japan why is'nt there something like Fritz and ChessBase? Or does it exist and we are living in a completly other Go-world? Chrilly P.S.: I do not want to offend anyone in this list. Everybody here does his best. I am just feed up with the things as they are. ___ 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/
Re: [computer-go] Re: 9x9 games wanted
Sil wrote: How about http://home.wwgo.jp/jp/minigo/ It seems that only 24 games are available. Is the whole collection available somewhere? Rémi ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
