Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
Hideki, Thank you. Your results look quite compelling. Do you allow memory (the number of nodes in the tree) to grow along with thinking time or is there a fixed limit? IIRC Don et. al.'s excellent scaling studies included gnugo but its effect was probably small. Self play dominated. Perhaps, what David Doshay calls, the evil twin effect causes self play to give the appearance of scaling better. - Dave Hillis -Original Message- From: Hideki Kato hideki_ka...@ybb.ne.jp To: computer-go computer-go@computer-go.org Sent: Sat, Oct 31, 2009 10:39 pm Subject: Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9). hillism...@netscape.net: 8cc26e08cfc0f77-5fd0-a...@webmail-m052.sysops.aol.com: -Original Message- From: Hideki Kato hideki_ka...@ybb.ne.jp To: computer-go computer-go@computer-go.org Sent: Wed, Oct 28, 2009 1:41 am Subject: Re: [computer-go] First ever win of a computer against a pro 9P as lack (game of Go, 9x9). ... BTW, recently I've measured the strength (win rate) vs time for a move curves with Zen vs GNU Go and Zen vs Zen (self-play) on 19 x 19 board. Without opening book, it saturates between +400 and +500 Elo against GNU but doesn't upto +800 Elo in self-play. That's somewhat interesting (detail will be open soon at GPW-2009). Hideki I'd say that is more than somewhat interesting. While we're waiting for the aper, can you give us a picture of how many games against Gnugo went into this nalysis? Do you see this in 9x9? I've attatched two charts of current results for convinience. hart1 is against GNU Go and Chart2 is self-play. The numbers for the 1st curve HA8000 (AMD Opteron 2.3GHz) 16 thread n Chart1 are: ime(s) Win DrawAll Dup WR std-dev Elo .02325 27 2,933 0 11.54% 0.59% -353.8 .1 509 23 728 0 71.50% 1.67% +159.8 .2 946 47 1,147 0 84.52% 1.07% +294.9 .5 1,803 60 2,000 0 91.65% 0.62% +416.2 .0 1,849 33 2,000 0 93.28% 0.56% +456.8 .0 4,455 121 4,812 0 93.84% 0.35% +473.1 The numbers for Chart2 are: ime(s) Win DrawAll Dup WR std-dev Elo .1 147 4 2,000 0 7.45% 0.59% -437.7 .3 992 36 2,000 0 50.50% 1.12% +3.5 .0 3,742 38 4,000 0 94.03% 0.37% +478.8 .0 13,157 43 13,328 1 98.89% 0.09% +779.3 Since above results are measured with no opening book, I'm now enchmarking opening book enabled but right now the samples are ot enough (642 games; see the 4th curve in Chart1, HA8000 (AMD pteron 2.3GHz) 16 thread w/ Book). Not a curve but a point now :-) For 9x9 it's not clear. The curve starts saturating near +500 Elo ut still seems increasing. Hideki - g...@nue.ci.i.u-tokyo.ac.jp (Kato) ___ omputer-go mailing list omputer...@computer-go.org ttp://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] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
dhillism...@netscape.net: 8cc28baed6fbe16-3fc0-16...@webmail-d068.sysops.aol.com: Hideki, Thank you. Your results look quite compelling. Do you allow memory (the number of nodes in the tree) to grow along with thinking time or is there a fixed limit? Each node of HA8000 cluster has 32 GB RAM which I believed is enough for a game with those time settings, up to 2s for a move, with no pondering, on 19x19. I observed, however, GC eventually run. I guess that affects little but I'll check it in the future experiments. IIRC Don et. al.'s excellent scaling studies included gnugo but its effect was probably small. Self play dominated. Perhaps, what David Doshay calls, the evil twin effect causes self play to give the appearance of scaling better. I have the same thought now. Perhaps my experimental results support such recent claim by strong players that strongest programs such as Zen are not so strong against human. It seems, however, too early to conclude anyway. Hidek -Original Message- From: Hideki Kato hideki_ka...@ybb.ne.jp To: computer-go computer-go@computer-go.org Sent: Sat, Oct 31, 2009 10:39 pm Subject: Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9). hillism...@netscape.net: 8cc26e08cfc0f77-5fd0-a...@webmail-m052.sysops.aol.com: -Original Message- From: Hideki Kato hideki_ka...@ybb.ne.jp To: computer-go computer-go@computer-go.org Sent: Wed, Oct 28, 2009 1:41 am Subject: Re: [computer-go] First ever win of a computer against a pro 9P as lack (game of Go, 9x9). ... BTW, recently I've measured the strength (win rate) vs time for a move curves with Zen vs GNU Go and Zen vs Zen (self-play) on 19 x 19 board. Without opening book, it saturates between +400 and +500 Elo against GNU but doesn't upto +800 Elo in self-play. That's somewhat interesting (detail will be open soon at GPW-2009). Hideki I'd say that is more than somewhat interesting. While we're waiting for the aper, can you give us a picture of how many games against Gnugo went into this nalysis? Do you see this in 9x9? I've attatched two charts of current results for convinience. hart1 is against GNU Go and Chart2 is self-play. The numbers for the 1st curve HA8000 (AMD Opteron 2.3GHz) 16 thread n Chart1 are: ime(s) Win DrawAll Dup WR std-dev Elo .02325 27 2,933 0 11.54% 0.59% -353.8 .1 509 23 728 0 71.50% 1.67% +159.8 .2 946 47 1,147 0 84.52% 1.07% +294.9 .5 1,803 60 2,000 0 91.65% 0.62% +416.2 .0 1,849 33 2,000 0 93.28% 0.56% +456.8 .0 4,455 121 4,812 0 93.84% 0.35% +473.1 The numbers for Chart2 are: ime(s) Win DrawAll Dup WR std-dev Elo .1 147 4 2,000 0 7.45% 0.59% -437.7 .3 992 36 2,000 0 50.50% 1.12% +3.5 .0 3,742 38 4,000 0 94.03% 0.37% +478.8 .0 13,157 43 13,328 1 98.89% 0.09% +779.3 Since above results are measured with no opening book, I'm now enchmarking opening book enabled but right now the samples are ot enough (642 games; see the 4th curve in Chart1, HA8000 (AMD pteron 2.3GHz) 16 thread w/ Book). Not a curve but a point now :-) For 9x9 it's not clear. The curve starts saturating near +500 Elo ut still seems increasing. Hideki - g...@nue.ci.i.u-tokyo.ac.jp (Kato) ___ omputer-go mailing list omputer...@computer-go.org ttp://www.computer-go.org/mailman/listinfo/computer-go/ inline file ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/ -- g...@nue.ci.i.u-tokyo.ac.jp (Kato) ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [SPAM] Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
2009/10/30 terry mcintyre terrymcint...@yahoo.com: This may be useful in computer Go. One of the reasons human pros do well is that they compute certain sub-problems once, and don't repeat the effort until something important changes. They know in an instant that certain positions are live or dead or seki; they know when a move ( reducing a liberty, for example ) disturbs that result. This could probably be emulated with theorem-proving ability. Presently, search algorithms have to rediscover these results many times over; this is (in my opinion) why computer programs get significantly weaker when starved for time; they cannot think deeply enough to solve problems which may be solved in an eyeblink by a pro. This sounds a lot like a description of GNU Go's persistent reading cache, which calculates reading shadow for all its readings. Has something similar tried for other programs? -- Seo Sanghyeon ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [SPAM] Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
Yes, this group does not have a consensus at all on this. On the one hand we hear that MCTS has reached a dead end and there is no benefit from extra CPU power, and on the other hand we have these developers hustling around for the biggest machines they can muster in order to play matches with humans! And Olivier claims that computers benefit more from additional thinking time than humans! Thanks for this comment. I agree that something is strange here :-) Olivier ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
2009/10/26 Don Dailey dailey@gmail.com: ... On the one hand we hear that MCTS has reached a dead end and there is no benefit from extra CPU power... Just curious, who actually claimed that and what was it based on? Erik ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [SPAM] Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
Just curious, who actually claimed that and what was it based on? I don't know who claimed it first, and who agreed for it, but I agree with it :-) More precisely, I think that increasing time and computational power makes computers stronger, but not for some particular things like long-term life-and-death in corners, or semeai situations. This makes a big limitation on what is possible with MCTS algorithms, in particular against humans. We made a lot of efforts for online learning of Monte-Carlo simulations, in order to improve this, but there's no significant improvement around that. Best regards, Olivier ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [SPAM] Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
2009/10/29 Olivier Teytaud olivier.teyt...@lri.fr Yes, this group does not have a consensus at all on this. On the one hand we hear that MCTS has reached a dead end and there is no benefit from extra CPU power, and on the other hand we have these developers hustling around for the biggest machines they can muster in order to play matches with humans! And Olivier claims that computers benefit more from additional thinking time than humans! Thanks for this comment. I agree that something is strange here :-) Olivier I'm being a bit sarcastic - I recognize that most of the statements made about this general issue are not based as much on logic as they are on emotional feelings or just making rash interpretations of tiny data samples. Almost always with us humans (myself included) when we try to interpret data we lean way in the direction of our own subjective biases. Even our own interpretations are in conflict sometimes.I myself have said things that upon close examination prove to be in conflict with something else I believed, they both could not be true! And then to add insult to injury we try to explain the conflict away with amazingly creative skill instead of just admitting that we might need to adjust our belief system. As far as this subject is concerned, I honestly don't think we fully understand it, myself included. We have a lot of conflicting evidence and I'm going to take a step backwards until we know more. With go it is extremely frustrating. The evaluation (rating/ranking) system is non-standard and rather kludgey and we would need thousands of games to settle this under controlled conditions unless the man/machine difference was enormous. - Don ___ 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: [SPAM] Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
2009/10/29 Olivier Teytaud olivier.teyt...@lri.fr Just curious, who actually claimed that and what was it based on? I don't know who claimed it first, and who agreed for it, but I agree with it :-) But you always seek the most hardware when you play against a human it seems. I think you realize it does help a lot to do this, otherwise your team would not be so foolish as to procure the big iron when it comes time to compete. You also are painfully aware that there are problems to be solved that will not easily succumb to just a few more doublings in power. That is exactly as it should be and is not a barrier. I don't think you know the difference between a wall and a point that is just far away. More precisely, I think that increasing time and computational power makes computers stronger, but not for some particular things like long-term life-and-death in corners, or semeai situations. This makes a big limitation on what is possible with MCTS algorithms, in particular against humans. We made a lot of efforts for online learning of Monte-Carlo simulations, in order to improve this, but there's no significant improvement around that. You are thinking with a very limited perspective here. Think in terms of 2 or 3 decades of Moores Law.We had those same barriers in chess that people said were impossible because we usually don't think in terms of getting 10,000 X more computing power, we are stuck in the present and just realize that getting 10X more is not nearly enough to solve some problem as you are observing here.And if 2 decades are not enough wait 2 more. I hope no one responds about Moores Law not holding any longer. That has nothing to do with my argument.My argument is that it takes a huge amount of extra CPU power to make a dent in big problems, just like it was in chess. No big surprise here.If Moores law doesn't hold then we are in trouble and it will take about twice as long. Why twice? I don't really know but by analogy the progress in chess software has been on par or slightly greater than the advances in hardware. (Most people don't realize this and think chess is 95% about hardware, but that is a complete misconception. In very rough terms there has been about the same increase in ELO due to software as to hardware over the last several years.) The combination of software and hardware is the potent combination if Moore's law will hold out for us.Just because it may not happen within the next 2 or 3 years doesn't mean it's a wall or that anything odd is going on here. - Don Best regards, Olivier ___ 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: [SPAM] Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
On Thu, Oct 29, 2009 at 12:00:32PM -0400, Don Dailey wrote: That is exactly as it should be and is not a barrier. I don't think you know the difference between a wall and a point that is just far away. I'd phrase this positively - the point is extremely far away with the current way MCTS will succumb to blunders because of the way it is completely unable to compensate for systematic bias (the amount of computation required to overcome the bias is extreme), but some clever algorithmic improvement could put the point much closer. This is just a discussion how steep a slope we will already call a wall, I think it's more productive to talk about how to make the slope less steep. :) -- Petr Pasky Baudis A lot of people have my books on their bookshelves. That's the problem, they need to read them. -- Don Knuth ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [SPAM] Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
On Thu, Oct 29, 2009 at 12:40 PM, Petr Baudis pa...@ucw.cz wrote: On Thu, Oct 29, 2009 at 12:00:32PM -0400, Don Dailey wrote: That is exactly as it should be and is not a barrier. I don't think you know the difference between a wall and a point that is just far away. I'd phrase this positively - the point is extremely far away with the current way MCTS will succumb to blunders because of the way it is completely unable to compensate for systematic bias (the amount of computation required to overcome the bias is extreme), but some clever algorithmic improvement could put the point much closer. This is just a discussion how steep a slope we will already call a wall, I think it's more productive to talk about how to make the slope less steep. :) I don't see it as a slope at all, just a matter of distance. So to me it's just a matter of continuing to put one foot in front of the other. But using different terminology than you, we should talk about how to get closer faster. As software people we have to attack it from the software end and not worry about the hardware end so much because that is out of our hands anyway (unless of course we are in hardware.) - Don -- Petr Pasky Baudis A lot of people have my books on their bookshelves. That's the problem, they need to read them. -- Don Knuth ___ 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: [SPAM] Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
Roger Penrose thinks the human brain can do things a Turing machine cannot. (Note: I don't say 'computer'.) He claims it's due to some quantum-physical effects used by the brain. I doubt his ideas are correct, but he did have a few interesting chess-positions to support his theory. Typically, they would contain a completely locked position, say a V-shaped pawn position and bishops on the wrong color to pass the pawn-ranks. These types of positions are very easily analyzed by even mediocre players, yet a computer never gets the right answer. Basically what it shows is that the human brain is able to conceptualize certain things that enable it to reason about situations that cannot be calculated by brute force. I don't claim that a Turing machine cannot do such things as well if programmed well, but it's very easy to see that there could be barriers to computers, no matter how much computing power you give them, if they solely rely on a simple method with brute force. Mark ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [SPAM] Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
That sounds to me like a dumb human with a smart algorithm can beat a fast computer with a dumb algorithm -- which speaks more to Penrose's reluctance to improve algorithms in his dumbed-down computer models than it does to any quantum-physical effects. Stir in some theorem-proving ability - where a great deal of research was accomplished decades ago - and a computer chess program can prove theorems about chess positions, including these bishops can never get past these pawns. This may be useful in computer Go. One of the reasons human pros do well is that they compute certain sub-problems once, and don't repeat the effort until something important changes. They know in an instant that certain positions are live or dead or seki; they know when a move ( reducing a liberty, for example ) disturbs that result. This could probably be emulated with theorem-proving ability. Presently, search algorithms have to rediscover these results many times over; this is (in my opinion) why computer programs get significantly weaker when starved for time; they cannot think deeply enough to solve problems which may be solved in an eyeblink by a pro. I've observed some high-dan-level amateurs playing complex semeai on 19x19 games. They might not actually know the result of a semeai, but they respond quickly to moves which would alter the status - if one of my liberties is taken, I take one of his - until such point as the player takes a noticeably long time to re-analyse the semeai and think I need not respond to that move and takes sente. The stronger the player, the more accurate these assessments are. Terry McIntyre terrymcint...@yahoo.com And one sad servitude alike denotes The slave that labours and the slave that votes -- Peter Pindar From: Mark Boon tesujisoftw...@gmail.com To: computer-go computer-go@computer-go.org Sent: Thu, October 29, 2009 10:14:18 AM Subject: Re: [SPAM] Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9). Roger Penrose thinks the human brain can do things a Turing machine cannot. (Note: I don't say 'computer'.) He claims it's due to some quantum-physical effects used by the brain. I doubt his ideas are correct, but he did have a few interesting chess-positions to support his theory. Typically, they would contain a completely locked position, say a V-shaped pawn position and bishops on the wrong color to pass the pawn-ranks. These types of positions are very easily analyzed by even mediocre players, yet a computer never gets the right answer. Basically what it shows is that the human brain is able to conceptualize certain things that enable it to reason about situations that cannot be calculated by brute force. I don't claim that a Turing machine cannot do such things as well if programmed well, but it's very easy to see that there could be barriers to computers, no matter how much computing power you give them, if they solely rely on a simple method with brute force. Mark ___ 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: [SPAM] Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
Yes, I agree with you on most of this. However, I believe that Go is a very simple domain in some sense and that we romanticize it too much. I am not saying there is not amazing depth to it, but it's represented very compactly and it's a game of perfect information with very limited choices. Having said that, I do fully appreciate that even if Moores law could hold indefinitely, there are still problems that will take decades to overcome if there are no software advances. - Don On Thu, Oct 29, 2009 at 1:14 PM, Mark Boon tesujisoftw...@gmail.com wrote: Roger Penrose thinks the human brain can do things a Turing machine cannot. (Note: I don't say 'computer'.) He claims it's due to some quantum-physical effects used by the brain. I doubt his ideas are correct, but he did have a few interesting chess-positions to support his theory. Typically, they would contain a completely locked position, say a V-shaped pawn position and bishops on the wrong color to pass the pawn-ranks. These types of positions are very easily analyzed by even mediocre players, yet a computer never gets the right answer. Basically what it shows is that the human brain is able to conceptualize certain things that enable it to reason about situations that cannot be calculated by brute force. I don't claim that a Turing machine cannot do such things as well if programmed well, but it's very easy to see that there could be barriers to computers, no matter how much computing power you give them, if they solely rely on a simple method with brute force. Mark ___ 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] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
-Original Message- From: Hideki Kato hideki_ka...@ybb.ne.jp To: computer-go computer-go@computer-go.org Sent: Wed, Oct 28, 2009 1:41 am Subject: Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9). ... BTW, recently I've measured the strength (win rate) vs time for a move curves with Zen vs GNU Go and Zen vs Zen (self-play) on 19 x 19 board. Without opening book, it saturates between +400 and +500 Elo against GNU but doesn't upto +800 Elo in self-play. That's somewhat interesting (detail will be open soon at GPW-2009). Hideki I'd say that is more than somewhat interesting. While we're waiting for the paper, can you give us a picture of how many games against Gnugo went into this analysis? Do you see this in 9x9? - Dave Hillis ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [SPAM] Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
What is interesting is not the fact that intrasitivity exists, that is not in doubt. But it quite interesting that this much intransitivity can be created with non-trivial and strong programs. I would like to see the data though, specifically the number of games between each player at each level and of course the scores that go with this. Such a differece indicates to me that the program (or MC programs in general) may be too brittle and needs some knowledge that gnuo has. - Don 2009/10/29 Olivier Teytaud olivier.teyt...@lri.fr BTW, recently I've measured the strength (win rate) vs time for a move curves with Zen vs GNU Go and Zen vs Zen (self-play) on 19 x 19 board. Without opening book, it saturates between +400 and +500 Elo against GNU but doesn't upto +800 Elo in self-play. That's somewhat interesting (detail will be open soon at GPW-2009). Just a post to say that I find this remark extremely interesting :-) Thanks a lot Hideki. Olivier ___ 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] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
dhillism...@netscape.net: 8cc26e08cfc0f77-5fd0-a...@webmail-m052.sysops.aol.com: -Original Message- From: Hideki Kato hideki_ka...@ybb.ne.jp To: computer-go computer-go@computer-go.org Sent: Wed, Oct 28, 2009 1:41 am Subject: Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9). ... BTW, recently I've measured the strength (win rate) vs time for a move curves with Zen vs GNU Go and Zen vs Zen (self-play) on 19 x 19 board. Without opening book, it saturates between +400 and +500 Elo against GNU but doesn't upto +800 Elo in self-play. That's somewhat interesting (detail will be open soon at GPW-2009). Hideki I'd say that is more than somewhat interesting. While we're waiting for the paper, can you give us a picture of how many games against Gnugo went into this analysis? Do you see this in 9x9? I'll post those pictures after back to Japan. The numbers of gamse are large enough (about 2,000 games around 50% of WR and up to 10,000 games at high WR) and it's on 19 x 19. Hideki -- g...@nue.ci.i.u-tokyo.ac.jp (Kato) ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
On Oct 27, 2009, at 7:41 PM, Hideki Kato wrote: IMHO, Jeff's idea is still very interesting while the implementation by the staff in Numenta have been going to not right direction. That was also my opinion. What I thought was strange is that Numenta's implementation doesn't have feed-back connections, which is a corner- stone of the ideas in the book. Those playouts are done in Cerebellum using some associative memory, I beleive. Then the mechanism, how to communicate with Cerebral, is a mistery, assuming some kind of tree search is done in Cerebral. It's not so sure to me there's a clear boundary between the activity of the two. It seems the tree search is done in the Cerebral cortex. But that may simply be because we're conscious of it. It's unclear what exactly happens during the unconscious processes. It mays also be a form of tree search that blends in with the conscious process. Knowledge about how the brain works is growing, but I believe it's mostly still a mystery. The way it's being observed currently is mostly like trying to figure out a computer-program by observing a piece of computer-memory on the screen. You see bits flashing on and off but you have to guess what instructed it to do so. The games in last Meijin-sen in Japan, Iyama vs Cho, may support your thought. I'm rather out of touch with what happens in tournaments. I've never heard of Iyama and even Cho could be a different one than I know. What happened in that match that is relevant to this discussion? Mark ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [SPAM] Re: [SPAM] Re: [SPAM] Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
But is it shown that the score is well done for these properties to hold in case of RAVE-guided exploration? Since it massively perpetuates any kind of MC bias... This only matters for the fact that we don't visit all the tree. For the consistency (the fact that asymptotically we will find the best possible decision), there's no problem. If score ~ success rate for n-- infinity (which holds for most usual rules, including rave rules) we also have that, for binary games, we have some good properties on the part of the tree which is visited. Please not that I do not claim that major improvements are possible in computer-go thanks to this. We only observed a very small improvement on success rates, and a good behavior on the situation which appeared during the game against Fan Hui. It might be interesting to know, for people who have similar problems in their bot (a situation in which, even with huge computation time, the good estimate is not found), they solve it with this. We use a stupid method, i.e. the success rate. The pattenrs are bigger than 3x3, with jokers in them. Bandits (Bernstein races, slightly modified) are used for distributing the computational effort among the tested patterns. Thank you for pointing me to more study material. :-) The following paper is great for Bernstein races: http://icml2008.cs.helsinki.fi/papers/523.pdf Please note, however, that we had only very small improvements with races. Maybe our code has had too many tuning steps in the past for being strongly improved by random generation of patterns and bernstein races for validating them. Best regards, Olivier ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [SPAM] Re: [SPAM] Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
If there are people interested in a ph.D. or a post-doc around Monte-Carlo Tree Search, candidates are welcome (Monte-Carlo Tree Search, and not necessarily / not only computer-go). Excuse me, but what press conference and where to ask? People interested in a ph.D. or a post doc can contact me. This was during a press conference at Taipei around a French-Taiwanese grant for joint research. but I can find no links even with Google. I'll ask to the taiwanese people if there is something on the web about the press conference. I was only there through a video. I don't know if there is something on the web. This is essentially for the launching of a France/Taiwan collaboration around Monte-Carlo Tree Search, I guess there are not thousards of journalists from tenths of countries interested in it :-) Best regards, Olivier ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
Mark Boon: 66913149-592c-426d-b52d-f52f3fa51...@gmail.com: On Oct 27, 2009, at 7:41 PM, Hideki Kato wrote: IMHO, Jeff's idea is still very interesting while the implementation by the staff in Numenta have been going to not right direction. That was also my opinion. What I thought was strange is that Numenta's implementation doesn't have feed-back connections, which is a corner- stone of the ideas in the book. Oh, I forgot to mention that, sorry. The feedback between layers in Cerebral cortex, which handle time I believe, is essential for the function of Cerebral and thus human, anyway. Those playouts are done in Cerebellum using some associative memory, I beleive. Then the mechanism, how to communicate with Cerebral, is a mistery, assuming some kind of tree search is done in Cerebral. It's not so sure to me there's a clear boundary between the activity of the two. It seems the tree search is done in the Cerebral cortex. But that may simply be because we're conscious of it. It's unclear what exactly happens during the unconscious processes. It mays also be a form of tree search that blends in with the conscious process. Knowledge about how the brain works is growing, but I believe it's mostly still a mystery. The way it's being observed currently is mostly like trying to figure out a computer-program by observing a piece of computer-memory on the screen. You see bits flashing on and off but you have to guess what instructed it to do so. Unluckily we have to have some strong assumption to analyze and mimic brain right now... My assumption is based on the experimental fact that the blood activity of Cerebral cortex of the professional players in both Shogi and Go increases a lot when reading forward positions. The games in last Meijin-sen in Japan, Iyama vs Cho, may support your thought. I'm rather out of touch with what happens in tournaments. I've never heard of Iyama and even Cho could be a different one than I know. What happened in that match that is relevant to this discussion? The games were very complicated and their thought was so deep and wide that the other professionals in another room in the venue couldn't follow nor understand. Hummm, I'm sorry but it's very difficult to explain my idea as it's rather some intution than logical thought, with non-mother language in addition. Hideki -- g...@nue.ci.i.u-tokyo.ac.jp (Kato) ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
Am I remembering correctly (maybe not) that Mogo communicates between nodes three times per second? That isn't a lot of communication opportunities if each turn lasts a few seconds. Olivier, have you tested parallel Mogo's ability to scale with core count at blitz speeds? I might imagine, for example, playing a series against itself with pondering turned off and one side playing blitz with 100 cores, and the other side playing with 10 cores each given 5 times as much thinking time. As others have said, congratulations... ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [SPAM] Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
Could you give us at least a general picture of improvements compared to what was last published as www.lri.fr/~teytaud/eg.pdfhttp://www.lri.fr/%7Eteytaud/eg.pdf? Is it just further tuning and small tweaks or are you trying out some exciting new things? ;-) There is one important improvement, for which I must check with coauthors if they agree for me to explain it. Below the other recent improvements in 9x9. We have also recently encoded some (very simple) tricks against bad cases as we had against Fan Hui (i.e. cases in which the only good move is not simulated). Roughly, is the value of the node is very bad, then simulate randomly among the sons. We can show (mathematically) that with such tricks, we have the consistency (as UCT), plus some frugality (i.e. we do not simulate all the tree, even with infinite computation time whereas UCT simulates all the tree AND simulates all the tree infinitely often). It gives very little improvement in self-play, but it understands better at least the situation seen in the game with Fan Hui. What I like in this improvement is that it's the first time there is something which was mathematically developped for mogo and which leads to a positive result. Well, maybe this changes only 1% of games, but maybe it makes mogo more robust for complicated ko fights which do not occur in self-play. Finally, there was a GP-based development of new patterns. However, this is quite minor I guess - I like the fact that this GP-based module works in a somehow stable manner, but maybe it would only be worth using it on an implementation which is not yet optimized. Best regards, Olivier ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [SPAM] Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
On Tue, Oct 27, 2009 at 08:47:41AM +0100, Olivier Teytaud wrote: Could you give us at least a general picture of improvements compared to what was last published as www.lri.fr/~teytaud/eg.pdfhttp://www.lri.fr/%7Eteytaud/eg.pdf? Is it just further tuning and small tweaks or are you trying out some exciting new things? ;-) There is one important improvement, for which I must check with coauthors if they agree for me to explain it. Below the other recent improvements in 9x9. We have also recently encoded some (very simple) tricks against bad cases as we had against Fan Hui (i.e. cases in which the only good move is not simulated). Roughly, is the value of the node is very bad, then simulate randomly among the sons. We can show (mathematically) that with such tricks, we have the consistency (as UCT), plus some frugality (i.e. we do not simulate all the tree, even with infinite computation time whereas UCT simulates all the tree AND simulates all the tree infinitely often). It gives very little improvement in self-play, but it understands better at least the situation seen in the game with Fan Hui. What I like in this improvement is that it's the first time there is something which was mathematically developped for mogo and which leads to a positive result. Well, maybe this changes only 1% of games, but maybe it makes mogo more robust for complicated ko fights which do not occur in self-play. Interesting! Conceptually, I don't like this that much since it just work-arounds RAVE bias instead of solving it in more general way, but I can see its technical value. AIUI, once upon N simulations in a node you take let's say the node with the lowest value, pick one son of it at random within the tree and start a simulation? Finally, there was a GP-based development of new patterns. However, this is quite minor I guess - I like the fact that this GP-based module works in a somehow stable manner, but maybe it would only be worth using it on an implementation which is not yet optimized. Wow - one of my planned little projects was genetic development of the 3x3 patterns... To evaluate patterns, do you use tournaments or some smarter method? I feared one generation would take awfully long... -- Petr Pasky Baudis A lot of people have my books on their bookshelves. That's the problem, they need to read them. -- Don Knuth ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
I suspect I am in your camp, Mark, though obviously it would be nice if we had measurements on this instead of conjectures. I will offer some anecdotal evidence concerning humans playing other humans, from club and tournament playing experience: you will find that shorter time limits amplify the winning probability of stronger players when humans play other humans. Beating somebody 2 stones stronger than you on Blitz is much harder than beating them on a longer time limit; you may find that you need 3 handicap stones. The bigger the strength difference, the worse it gets. Beating a professional player in Blitz Go is *ferociously* difficult, even with very high handicap. Humans are extremely good at recognizing patterns, whole board awareness, and intuition about influence; reading more into the game is useless for a human without these skills. It has never really surprised me that stronger players are that much better at short time limits, given their larger experience and knowledge. At the extreme end, you have the beginner: even if it's a human with incredible reading ability, he/she will still lose with a three hour time limit, to somebody a few stones stronger, and on a 10 minute limit. Well, some trials with different time limits against computers would be nice, I guess :) Christian On 26/10/2009 23:14, Mark Boon wrote: 2009/10/26 Don Daileydailey@gmail.com: Yes, you understood me right. I disagree with Olivier on this one.To me it is self-evident that humans are more scalable than computers because we have better heuristics. When that is not true it is usually because the task is trivial, not because it is hard. Personally I rather think that what makes a human good at certain tasks is not necessarily a conscious effort, and that doesn't have to be a trivial task. So then actively thinking longer doesn't help as much because you lack the control over the thought-process. I believe very much that Go falls in that category, where Chess does not. Mark ___ 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] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
I will offer some anecdotal evidence concerning humans playing other humans, from club and tournament playing experience: you will find that shorter time limits amplify the winning probability of stronger players... Another anecdote. At a Fost Cup (Computer Go tournament) from 10-15 years ago, a pro player had made his own program. I think it was based on patterns and though it wasn't one of the stronger programs, it played very quickly. This was at Nihon Kiin, and another pro friend popped in to visit; I forget his name, but he was one of the top 9p players. He played the fast program, and they played a 19x19 game at the pace of at least 60 moves/minute. I forget if it was an even game or 9-stone handicap as it didn't matter - the pro killed every group. But what impressed me was he made shapes and strength that even dan players would've had to work hard to get. A wall of stones along one side of the board naturally ended up being in just the right place to work with joseki played earlier on the other side of the board, stones played long before ended up on just the critical points to kill, yet he took not even a breath to plan any of this. So, I wonder if the blitz strength of very strong go players is something special and peculiar to the game of go. Patterns and shape knowledge is so important in go, that humans (*) gain relatively little extra strength from extra thinking. Darren *: Meaning very strong players who've spent years studying and appreciating good shape. -- Darren Cook, Software Researcher/Developer http://dcook.org/gobet/ (Shodan Go Bet - who will win?) http://dcook.org/mlsn/ (Multilingual open source semantic network) http://dcook.org/work/ (About me and my work) http://dcook.org/blogs.html (My blogs and articles) ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
I strongly believe that such patterns must not be only spatial (static) but also temporal, ie, dynamic or sequence of pattens which allow the player quickly remember the results of local fights or LD. Hideki Darren Cook: 4ae6d9b6.1070...@dcook.org: I will offer some anecdotal evidence concerning humans playing other humans, from club and tournament playing experience: you will find that shorter time limits amplify the winning probability of stronger players... Another anecdote. At a Fost Cup (Computer Go tournament) from 10-15 years ago, a pro player had made his own program. I think it was based on patterns and though it wasn't one of the stronger programs, it played very quickly. This was at Nihon Kiin, and another pro friend popped in to visit; I forget his name, but he was one of the top 9p players. He played the fast program, and they played a 19x19 game at the pace of at least 60 moves/minute. I forget if it was an even game or 9-stone handicap as it didn't matter - the pro killed every group. But what impressed me was he made shapes and strength that even dan players would've had to work hard to get. A wall of stones along one side of the board naturally ended up being in just the right place to work with joseki played earlier on the other side of the board, stones played long before ended up on just the critical points to kill, yet he took not even a breath to plan any of this. So, I wonder if the blitz strength of very strong go players is something special and peculiar to the game of go. Patterns and shape knowledge is so important in go, that humans (*) gain relatively little extra strength from extra thinking. Darren *: Meaning very strong players who've spent years studying and appreciating good shape. -- g...@nue.ci.i.u-tokyo.ac.jp (Kato) ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [SPAM] Re: [SPAM] Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
AIUI, once upon N simulations in a node you take let's say the node with the lowest value, pick one son of it at random within the tree and start a simulation? I'll try to write it clearly (for binary deterministic games, extensions can be shown but they are too long and out of topic in this mailing list I guess :-) ): If (average value of father threshold ) then randomly pick up one son else pick up the son with maximum score end If the score is asymptotically equivalent to the success rate, and if the threshold is 0 and 1, then this ensures consistency (convergence to optimal move). If the score is well done, then this is consistent without visiting all the tree. UCT (with non-zero constant) visits all the tree, and does so infinitely often. UCT (with zero constant) does not visit all the tree, but it is not necessarily consistent. Wow - one of my planned little projects was genetic development of the 3x3 patterns... To evaluate patterns, do you use tournaments or some smarter method? I feared one generation would take awfully long... We use a stupid method, i.e. the success rate. The pattenrs are bigger than 3x3, with jokers in them. Bandits (Bernstein races, slightly modified) are used for distributing the computational effort among the tested patterns. Best regards, Olivier ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
[computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
Dear all, some comments by my Taiwanese colleagues about the game played by MoGo against the 9p pro: 1) mogoTW finally ran on the 16*8 system on Oct. 26, 2009. 2) Contributors for which I did not know their real name: Hsien-Der Huang and Cheng-Wei Chou (sorry for them!) 3) Some comments by the pro, translated by people from Taiwan: === He told the jounalists that he made a mistake on move 26, which is a move that it is hard to find such kind an error (move 26) even for a professional Go player. However, he was so surprised that MoGoTW found this mistake that he made during the game. That means the level of MoGoTW in 9x9 game has made an improvement than last time (08/2009). In addition, he also told the journalists that MoGoTW made a good move at Move 29. He thought that the main reason that MoGoTW won the game was Move 29 and MoGoTW didn't make any mistake after Move 29. == ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
On Oct 27, 2009, at 3:39 AM, Hideki Kato wrote: I strongly believe that such patterns must not be only spatial (static) but also temporal, ie, dynamic or sequence of pattens which allow the player quickly remember the results of local fights or LD. I think that's exactly right. At least for humans. Maybe for computers there's another way. After reading On Intelligence (anyone follow my advice and read it?) I got to thinking the human brain possibly does a lot of little playouts in parallel. Not random, whole-board playouts from beginning to end, but short, local playouts, following strong patterns at each choice. Each time the result is fed back into the first layer so that the result of this playout gets used to guide the next playout. And the variance of the outcome of each of these playouts gets fed into the next layer to recognise higher-level concepts. Maybe for a few levels until it reaches a conscious level. The reason why thinking longer only helps marginally is that these small playouts follow a limited set of patterns. It takes time and practice to add these patterns, you can't easily consciously add a pattern in there during the game. So 'thinking' is restricted to a higher level, trying to think steps ahead in the game. Obviously this helps a lot for strength too, and pros are very good at that too. But with each stone you read ahead it becomes harder for your brain to do the pattern-matching because it doesn't have the complete (visual) input. So humans tend to think ahead in rather fixed sequences along the lines of play in the patterns that are followed sub-consciously. So when Sakata claimed he can read ahead 30 moves in a blink, he doesn't do a search of lots of possibilities. Instead, his brain is able to do these little playouts a lot deeper than mere mortals can. Most likely the main candidates all come up in the first (split) second. The rest of the time he spends verifying their results. This is all rather speculative of course. Christian Nentwich is right that it would be nice if we could measure this somehow. That's going to be difficult. But it shows a bit why I have the opinion that I have. Mark ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [SPAM] Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
I forgot the most important thing around this win against a pro: this press conference was for the starting of a project, and in this project we have funding for ph.D. or postdocs. If there are people interested in a ph.D. or a post-doc around Monte-Carlo Tree Search, candidates are welcome (Monte-Carlo Tree Search, and not necessarily / not only computer-go). Best regards, Olivier ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [SPAM] Re: [SPAM] Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
On Tue, Oct 27, 2009 at 06:32:44PM +0200, Olivier Teytaud wrote: AIUI, once upon N simulations in a node you take let's say the node with the lowest value, pick one son of it at random within the tree and start a simulation? I'll try to write it clearly (for binary deterministic games, extensions can be shown but they are too long and out of topic in this mailing list I guess :-) ): If (average value of father threshold ) then randomly pick up one son else pick up the son with maximum score end Aha, thanks for clearing that up. If the score is asymptotically equivalent to the success rate, and if the threshold is 0 and 1, then this ensures consistency (convergence to optimal move). If the score is well done, then this is consistent without visiting all the tree. But is it shown that the score is well done for these properties to hold in case of RAVE-guided exploration? Since it massively perpetuates any kind of MC bias... Wow - one of my planned little projects was genetic development of the 3x3 patterns... To evaluate patterns, do you use tournaments or some smarter method? I feared one generation would take awfully long... We use a stupid method, i.e. the success rate. The pattenrs are bigger than 3x3, with jokers in them. Bandits (Bernstein races, slightly modified) are used for distributing the computational effort among the tested patterns. Thank you for pointing me to more study material. :-) -- Petr Pasky Baudis A lot of people have my books on their bookshelves. That's the problem, they need to read them. -- Don Knuth ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [SPAM] Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
Olivier Teytaud: aa5e3c330910271105ocd762e8xb283fd386f20b...@mail.gmail.com: I forgot the most important thing around this win against a pro: this press conference was for the starting of a project, and in this project we have funding for ph.D. or postdocs. If there are people interested in a ph.D. or a post-doc around Monte-Carlo Tree Search, candidates are welcome (Monte-Carlo Tree Search, and not necessarily / not only computer-go). Excuse me, but what press conference and where to ask? You wrote in your first post of this thread, This was during a press conference at Taipei around a French-Taiwanese grant for joint research. but I can find no links even with Google. Hideki -- g...@nue.ci.i.u-tokyo.ac.jp (Kato) ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
Mark Boon: 4ec4bc46-e52f-4ac2-a7ff-edaf17de3...@gmail.com: On Oct 27, 2009, at 3:39 AM, Hideki Kato wrote: I strongly believe that such patterns must not be only spatial (static) but also temporal, ie, dynamic or sequence of pattens which allow the player quickly remember the results of local fights or LD. I think that's exactly right. At least for humans. Maybe for computers there's another way. That could be a challenging problem in this century... After reading On Intelligence (anyone follow my advice and read it?) I bought (:-) English version and read Japanese one immediate after their publish. IMHO, Jeff's idea is still very interesting while the implementation by the staff in Numenta have been going to not right direction. More generalized version of his idea could be that Cerebral cortex works to reduce temporal error similar to visual cortex but spatial error. I got to thinking the human brain possibly does a lot of little playouts in parallel. Not random, whole-board playouts from beginning to end, but short, local playouts, following strong patterns at each choice. Each time the result is fed back into the first layer so that the result of this playout gets used to guide the next playout. And the variance of the outcome of each of these playouts gets fed into the next layer to recognise higher-level concepts. Maybe for a few levels until it reaches a conscious level. Those playouts are done in Cerebellum using some associative memory, I beleive. Then the mechanism, how to communicate with Cerebral, is a mistery, assuming some kind of tree search is done in Cerebral. The reason why thinking longer only helps marginally is that these small playouts follow a limited set of patterns. It takes time and practice to add these patterns, you can't easily consciously add a pattern in there during the game. Agree. So 'thinking' is restricted to a higher level, trying to think steps ahead in the game. Obviously this helps a lot for strength too, and pros are very good at that too. But with each stone you read ahead it becomes harder for your brain to do the pattern-matching because it doesn't have the complete (visual) input. So humans tend to think ahead in rather fixed sequences along the lines of play in the patterns that are followed sub-consciously. So when Sakata claimed he can read ahead 30 moves in a blink, he doesn't do a search of lots of possibilities. Instead, his brain is able to do these little playouts a lot deeper than mere mortals can. Most likely the main candidates all come up in the first (split) second. The rest of the time he spends verifying their results. The games in last Meijin-sen in Japan, Iyama vs Cho, may support your thought. This is all rather speculative of course. Christian Nentwich is right that it would be nice if we could measure this somehow. That's going to be difficult. But it shows a bit why I have the opinion that I have. BTW, recently I've measured the strength (win rate) vs time for a move curves with Zen vs GNU Go and Zen vs Zen (self-play) on 19 x 19 board. Without opening book, it saturates between +400 and +500 Elo against GNU but doesn't upto +800 Elo in self-play. That's somewhat interesting (detail will be open soon at GPW-2009). Hideki -- g...@nue.ci.i.u-tokyo.ac.jp (Kato) ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
[computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
Dear all, For information, our Taiwanese partners(**) for a ANR grant have organized public demonstration games between MoGoTW (based on MoGo 4.86.Soissons + the TW modifications developped jointly with our Taiwanese colleagues) and C.-H. Chou 9P, top pro player winner of the LG Cup 2007. This was during a press conference at Taipei around a French-Taiwanese grant for joint research. Details: a) MoGoTW was running on 32 quad-cores(*) in Taiwan. b) There were two blitz games (15 minutes per side), won by the pro. c) There was one non-blitz game (45 minutes per side). MoGo was unlucky as it was black, but it nonetheless won the game. This game is enclosed. All games can be found on KGS (account nutngo) Remarks: a) Fuego won as white against a 9P a few months ago. Therefore computers have won both as white and black against top players :-) We now should win on a complete game like 4 out of 7 games and the job would be completly done for 9x9 Go :-) b) MoGo already won a game as black, against Catalin Taranu, but I guess the pro, at that time, had played an original opening somehow for fun (I'm not sure of that, however). c) My feeling is that blitz games are not favorable to computers... Statistics are in accordance with this I guess. Humans are stronger for short time settings. d) If I understand well, MoGo won a final semeai in the upper right part. But, nearly everybody on this mailing (except you, Sylvain, maybe, if you still read this mailing-list :-) ?) reads go games better than me, so don't trust this comment :-) e) The game was longer than most important games I've seen (59 moves). All comments welcome. Best regards Olivier (*) mogoTW was supposed to run on this 32x4 system, but other platforms were prepared in case of trouble on this cluster. I'll publish a correction if I see that the game was not played on this machine. (**) contributors include all the mogo-people, plus Mei-Hui Wang, Chang-Shing Lee, Shi-Jim Yen, and people that I only know by their nicknames (Coldmilk, TomTom...) - sorry for the people I've forgotten, names in Chinese are difficult for me :-) 20091026-1-Zhou vs. MoGoTW9X9.sgf Description: application/go-sgf ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
How things changes. You would never hear a comment like Remark c) below concerning the old alpha-beta chess engines. Olivier Teytaud wrote: Dear all, For information, our Taiwanese partners(**) for a ANR grant have organized public demonstration games between MoGoTW (based on MoGo 4.86.Soissons + the TW modifications developped jointly with our Taiwanese colleagues) and C.-H. Chou 9P, top pro player winner of the LG Cup 2007. This was during a press conference at Taipei around a French-Taiwanese grant for joint research. Details: a) MoGoTW was running on 32 quad-cores(*) in Taiwan. b) There were two blitz games (15 minutes per side), won by the pro. c) There was one non-blitz game (45 minutes per side). MoGo was unlucky as it was black, but it nonetheless won the game. This game is enclosed. All games can be found on KGS (account nutngo) Remarks: a) Fuego won as white against a 9P a few months ago. Therefore computers have won both as white and black against top players :-) We now should win on a complete game like 4 out of 7 games and the job would be completly done for 9x9 Go :-) b) MoGo already won a game as black, against Catalin Taranu, but I guess the pro, at that time, had played an original opening somehow for fun (I'm not sure of that, however). c) My feeling is that blitz games are not favorable to computers... Statistics are in accordance with this I guess. Humans are stronger for short time settings. d) If I understand well, MoGo won a final semeai in the upper right part. But, nearly everybody on this mailing (except you, Sylvain, maybe, if you still read this mailing-list :-) ?) reads go games better than me, so don't trust this comment :-) e) The game was longer than most important games I've seen (59 moves). All comments welcome. Best regards Olivier (*) mogoTW was supposed to run on this 32x4 system, but other platforms were prepared in case of trouble on this cluster. I'll publish a correction if I see that the game was not played on this machine. (**) contributors include all the mogo-people, plus Mei-Hui Wang, Chang-Shing Lee, Shi-Jim Yen, and people that I only know by their nicknames (Coldmilk, TomTom...) - sorry for the people I've forgotten, names in Chinese are difficult for me :-) ___ 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] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
Hi! On Mon, Oct 26, 2009 at 07:19:45PM +0100, Olivier Teytaud wrote: For information, our Taiwanese partners(**) for a ANR grant have organized public demonstration games between Thanks for the information! MoGoTW (based on MoGo 4.86.Soissons + the TW modifications developped jointly with our Taiwanese colleagues) and C.-H. Chou 9P, top pro player winner of the LG Cup 2007. Could you give us at least a general picture of improvements compared to what was last published as www.lri.fr/~teytaud/eg.pdf ? Is it just further tuning and small tweaks or are you trying out some exciting new things? ;-) c) My feeling is that blitz games are not favorable to computers... Statistics are in accordance with this I guess. Humans are stronger for short time settings. Maybe in high-level 9x9 games that's true, but as a general statement I'd dispute this, at least in watching 5k-1k-level 19x19 MCTS games on KGS I got a completely different impression; humans are much more -- Petr Pasky Baudis A lot of people have my books on their bookshelves. That's the problem, they need to read them. -- Don Knuth ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
2009/10/26 Richard J. Lorentz lore...@csun.edu How things changes. You would never hear a comment like Remark c) below concerning the old alpha-beta chess engines. Yes, this group does not have a consensus at all on this. On the one hand we hear that MCTS has reached a dead end and there is no benefit from extra CPU power, and on the other hand we have these developers hustling around for the biggest machines they can muster in order to play matches with humans! And Olivier claims that computers benefit more from additional thinking time than humans! - Don Olivier Teytaud wrote: Dear all, For information, our Taiwanese partners(**) for a ANR grant have organized public demonstration games between MoGoTW (based on MoGo 4.86.Soissons + the TW modifications developped jointly with our Taiwanese colleagues) and C.-H. Chou 9P, top pro player winner of the LG Cup 2007. This was during a press conference at Taipei around a French-Taiwanese grant for joint research. Details: a) MoGoTW was running on 32 quad-cores(*) in Taiwan. b) There were two blitz games (15 minutes per side), won by the pro. c) There was one non-blitz game (45 minutes per side). MoGo was unlucky as it was black, but it nonetheless won the game. This game is enclosed. All games can be found on KGS (account nutngo) Remarks: a) Fuego won as white against a 9P a few months ago. Therefore computers have won both as white and black against top players :-) We now should win on a complete game like 4 out of 7 games and the job would be completly done for 9x9 Go :-) b) MoGo already won a game as black, against Catalin Taranu, but I guess the pro, at that time, had played an original opening somehow for fun (I'm not sure of that, however). c) My feeling is that blitz games are not favorable to computers... Statistics are in accordance with this I guess. Humans are stronger for short time settings. d) If I understand well, MoGo won a final semeai in the upper right part. But, nearly everybody on this mailing (except you, Sylvain, maybe, if you still read this mailing-list :-) ?) reads go games better than me, so don't trust this comment :-) e) The game was longer than most important games I've seen (59 moves). All comments welcome. Best regards Olivier (*) mogoTW was supposed to run on this 32x4 system, but other platforms were prepared in case of trouble on this cluster. I'll publish a correction if I see that the game was not played on this machine. (**) contributors include all the mogo-people, plus Mei-Hui Wang, Chang-Shing Lee, Shi-Jim Yen, and people that I only know by their nicknames (Coldmilk, TomTom...) - sorry for the people I've forgotten, names in Chinese are difficult for me :-) -- ___ computer-go mailing listcomputer...@computer-go.orghttp://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] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
Peter, did your comment get cut off? Anyway, I agree with you on this. Humans are not stronger on short time settings. I believe that SOME humans could be better if they have a problem staying interested for a longer period of time and the longer time control upsets their rhythm or something. But I don't believe it's a general rule. I did know a chess player who was a weak expert and all he did was play speed chess all day long. In tournaments with long time controls, he still played speed chess. It was crazy, finishing his games after only having used 5 or 10 minutes. He claimed that he did not need longer to think because he was always sure the move he played was the best. Of course this is completely ridiculous since he was hundreds of ELO below the best human players and even further from perfect play. - Don On Mon, Oct 26, 2009 at 3:58 PM, Petr Baudis pa...@ucw.cz wrote: Hi! On Mon, Oct 26, 2009 at 07:19:45PM +0100, Olivier Teytaud wrote: For information, our Taiwanese partners(**) for a ANR grant have organized public demonstration games between Thanks for the information! MoGoTW (based on MoGo 4.86.Soissons + the TW modifications developped jointly with our Taiwanese colleagues) and C.-H. Chou 9P, top pro player winner of the LG Cup 2007. Could you give us at least a general picture of improvements compared to what was last published as www.lri.fr/~teytaud/eg.pdfhttp://www.lri.fr/%7Eteytaud/eg.pdf? Is it just further tuning and small tweaks or are you trying out some exciting new things? ;-) c) My feeling is that blitz games are not favorable to computers... Statistics are in accordance with this I guess. Humans are stronger for short time settings. Maybe in high-level 9x9 games that's true, but as a general statement I'd dispute this, at least in watching 5k-1k-level 19x19 MCTS games on KGS I got a completely different impression; humans are much more -- Petr Pasky Baudis A lot of people have my books on their bookshelves. That's the problem, they need to read them. -- Don Knuth ___ 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] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
On Mon, Oct 26, 2009 at 04:20:24PM -0400, Don Dailey wrote: Peter, did your comment get cut off? Oops, indeed. Prone to tactical mistakes in high time pressure is what I meant to say. Anyway, I agree with you on this. Humans are not stronger on short time settings. I believe that SOME humans could be better if they have a problem staying interested for a longer period of time and the longer time control upsets their rhythm or something. But I don't believe it's a general rule. Well, of course most humans play better with more time, the question is whether they or the computer gain more from the extra time. And I think while between, let's say 30s/move and 10min/move the curve of such advantage could be pretty straight, I think it would behave quite differently at the extreme ends. -- Petr Pasky Baudis A lot of people have my books on their bookshelves. That's the problem, they need to read them. -- Don Knuth ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
[computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
Congratulations Olivier and the MoGo team! Good job! Now let us know the secrets of MoGoTW :) Did you get pro commentary on the game? Martin ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
2009/10/26 Don Dailey dailey@gmail.com: 2009/10/26 Richard J. Lorentz lore...@csun.edu Yes, this group does not have a consensus at all on this. On the one hand we hear that MCTS has reached a dead end and there is no benefit from extra CPU power, and on the other hand we have these developers hustling around for the biggest machines they can muster in order to play matches with humans! And Olivier claims that computers benefit more from additional thinking time than humans! Well, we had this discussion a while back on this list. I (and some others) argued that humans play fast extremely well and that more time provides a rapidly decreasing benefit. If I remember well it was you who was arguing this not being the case and that pros benefit greatly with more time. So it seems we're starting to see some support for the argument that at least in Go professional players don't benefit as much from more time than computers do at the moment. Mark ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
Yes, you understood me right. I disagree with Olivier on this one.To me it is self-evident that humans are more scalable than computers because we have better heuristics. When that is not true it is usually because the task is trivial, not because it is hard. - Don On Mon, Oct 26, 2009 at 6:14 PM, Mark Boon tesujisoftw...@gmail.com wrote: 2009/10/26 Don Dailey dailey@gmail.com: 2009/10/26 Richard J. Lorentz lore...@csun.edu Yes, this group does not have a consensus at all on this. On the one hand we hear that MCTS has reached a dead end and there is no benefit from extra CPU power, and on the other hand we have these developers hustling around for the biggest machines they can muster in order to play matches with humans! And Olivier claims that computers benefit more from additional thinking time than humans! Well, we had this discussion a while back on this list. I (and some others) argued that humans play fast extremely well and that more time provides a rapidly decreasing benefit. If I remember well it was you who was arguing this not being the case and that pros benefit greatly with more time. So it seems we're starting to see some support for the argument that at least in Go professional players don't benefit as much from more time than computers do at the moment. Mark ___ 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] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
2009/10/26 Don Dailey dailey@gmail.com: Yes, you understood me right. I disagree with Olivier on this one. To me it is self-evident that humans are more scalable than computers because we have better heuristics. When that is not true it is usually because the task is trivial, not because it is hard. Personally I rather think that what makes a human good at certain tasks is not necessarily a conscious effort, and that doesn't have to be a trivial task. So then actively thinking longer doesn't help as much because you lack the control over the thought-process. I believe very much that Go falls in that category, where Chess does not. Mark ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
RE: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9).
Congratulations. Can you put it on cgos 9x9 so we can see what cgos rating it takes to beat a pro? Maybe zen can return at the same time so we can get a comparison. David From: computer-go-boun...@computer-go.org [mailto:computer-go-boun...@computer-go.org] On Behalf Of Olivier Teytaud Sent: Monday, October 26, 2009 11:20 AM To: computer-go Subject: [computer-go] First ever win of a computer against a pro 9P as black (game of Go, 9x9). Dear all, For information, our Taiwanese partners(**) for a ANR grant have organized public demonstration games between MoGoTW (based on MoGo 4.86.Soissons + the TW modifications developped jointly with our Taiwanese colleagues) and C.-H. Chou 9P, top pro player winner of the LG Cup 2007. This was during a press conference at Taipei around a French-Taiwanese grant for joint research. Details: a) MoGoTW was running on 32 quad-cores(*) in Taiwan. b) There were two blitz games (15 minutes per side), won by the pro. c) There was one non-blitz game (45 minutes per side). MoGo was unlucky as it was black, but it nonetheless won the game. This game is enclosed. All games can be found on KGS (account nutngo) Remarks: a) Fuego won as white against a 9P a few months ago. Therefore computers have won both as white and black against top players :-) We now should win on a complete game like 4 out of 7 games and the job would be completly done for 9x9 Go :-) b) MoGo already won a game as black, against Catalin Taranu, but I guess the pro, at that time, had played an original opening somehow for fun (I'm not sure of that, however). c) My feeling is that blitz games are not favorable to computers... Statistics are in accordance with this I guess. Humans are stronger for short time settings. d) If I understand well, MoGo won a final semeai in the upper right part. But, nearly everybody on this mailing (except you, Sylvain, maybe, if you still read this mailing-list :-) ?) reads go games better than me, so don't trust this comment :-) e) The game was longer than most important games I've seen (59 moves). All comments welcome. Best regards Olivier (*) mogoTW was supposed to run on this 32x4 system, but other platforms were prepared in case of trouble on this cluster. I'll publish a correction if I see that the game was not played on this machine. (**) contributors include all the mogo-people, plus Mei-Hui Wang, Chang-Shing Lee, Shi-Jim Yen, and people that I only know by their nicknames (Coldmilk, TomTom...) - sorry for the people I've forgotten, names in Chinese are difficult for me :-) ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
