RE: [computer-go] Poll: how long until computers are as strong as pros?
Self play results are much better than play against another opponent (since the faster version sees everything the slower one does, plus more). At stronger levels, the win rate for a stone difference is higher. Pure computer power increase will take much longer than your estimate. On the other hand, monte carlo is still new, and there will be big improvements in the algorithm without more processing power. I think the algorithm is more important, so perhaps a top pro will lose an even 19x19 in 20 years or less. David -Original Message- From: computer-go-boun...@computer-go.org [mailto:computer-go- boun...@computer-go.org] On Behalf Of Bob Hearn Sent: Thursday, February 12, 2009 9:42 PM To: computer-go Subject: [computer-go] Poll: how long until computers are as strong as pros? How long until a computer beats a pro -- any pro -- in an even game? How long until a computer can routinely beat the best pros? Not a very scientific poll, I realize, but I'd like some numbers to use in my AAAS talk on Saturday. FWIW, this is a back-of-the-envelope calculation I did in August, when MoGo beat Myungwan Kim 8p at H9: After the match, one of the MoGo programmers mentioned that doubling the computation led to a 63% win rate against the baseline version, and that so far this scaling seemed to continue as computation power increased. So -- quick back-of-the-envelope calculation, tell me where I am wrong. 63% win rate = about half a stone advantage in go. So we need 4x processing power to increase by a stone. At the current rate of Moore's law, that's about 4 years. Kim estimated that the game with MoGo would be hard at 8 stones. That suggests that in 32 years a supercomputer comparable to the one that played in this match would be as strong as Kim. This calculation is optimistic in assuming that you can meaningfully scale the 63% win rate indefinitely, especially when measuring strength against other opponents, and not a weaker version of itself. It's also pessimistic in assuming there will be no improvement in the Monte Carlo technique. But still, 32 years seems like a surprisingly long time, much longer than the 10 years that seems intuitively reasonable. Naively, it would seem that improvements in the Monte Carlo algorithms could gain some small number of stones in strength for fixed computation, but that would just shrink the 32 years by maybe a decade. Thanks, Bob Hearn - Robert A. Hearn Neukom Institute for Computational Science, Dartmouth College robert.a.he...@dartmouth.edu http://www.dartmouth.edu/~rah/ ___ 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] Poll: how long until computers are as strong as pros?
I think this estimate is a reasonable educated guess. The uncertainties are quite big. I would say your estimate has a total margin of error of at least 50% (it will probably take between 15 years and 50 years) but I don't think it's possible to estimate much more accurate at this stage. Dave Van: computer-go-boun...@computer-go.org namens Bob Hearn Verzonden: vr 13-2-2009 6:42 Aan: computer-go Onderwerp: [computer-go] Poll: how long until computers are as strong as pros? How long until a computer beats a pro -- any pro -- in an even game? How long until a computer can routinely beat the best pros? Not a very scientific poll, I realize, but I'd like some numbers to use in my AAAS talk on Saturday. FWIW, this is a back-of-the-envelope calculation I did in August, when MoGo beat Myungwan Kim 8p at H9: After the match, one of the MoGo programmers mentioned that doubling the computation led to a 63% win rate against the baseline version, and that so far this scaling seemed to continue as computation power increased. So -- quick back-of-the-envelope calculation, tell me where I am wrong. 63% win rate = about half a stone advantage in go. So we need 4x processing power to increase by a stone. At the current rate of Moore's law, that's about 4 years. Kim estimated that the game with MoGo would be hard at 8 stones. That suggests that in 32 years a supercomputer comparable to the one that played in this match would be as strong as Kim. This calculation is optimistic in assuming that you can meaningfully scale the 63% win rate indefinitely, especially when measuring strength against other opponents, and not a weaker version of itself. It's also pessimistic in assuming there will be no improvement in the Monte Carlo technique. But still, 32 years seems like a surprisingly long time, much longer than the 10 years that seems intuitively reasonable. Naively, it would seem that improvements in the Monte Carlo algorithms could gain some small number of stones in strength for fixed computation, but that would just shrink the 32 years by maybe a decade. Thanks, Bob Hearn - Robert A. Hearn Neukom Institute for Computational Science, Dartmouth College robert.a.he...@dartmouth.edu http://www.dartmouth.edu/~rah/ ___ 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] Poll: how long until computers are as strong as pros?
From: Bob Hearn robert.a.he...@dartmouth.edu How long until a computer beats a pro -- any pro -- in an even game? How long until a computer can routinely beat the best pros? We've recently seen a program with a 7 stone handicap beat a pro, so we're a little bit closer than when you made those computations. I agree with David Fotland, there will be some serious algorithmic improvements; we are hardly scratching the surface at this point. In addition, the computing field in general has been held back for two decades by excessive dependence on the x86 family of architectures; this trend is going to change; performance-per-million-transisters is going to rise sharply in the next decade. Reconfigurable computing has yet to fulfill its aims, but a variety of technologies are likely to come together to make it possible for pro-level computer programs - and a lot of other goodness - within the next 20 years. Something like SGI's Molecule and the Sicortex architectures will make it possible for lots of low-power minimalist computers to be clustered together at comparatively low cost, using less rack space and power. Today's commodity-based clusters waste an amazing amount of hardware. Why should a 4000-node computer have 1000 VGA ports, 1000 disk drives, 1000 disk controllers, 1000 power supplies? We need to create commodity compute modules which are a lot leaner, smaller, cheaper, faster, and more efficient. A compute node should have one or many CPUs, memory controllers, fast inter-node communications, local flash (or a succesor thereof) for boot and other longer-term info, and nothing else - no video, no disk, no independent fans and power supplies, etc. It could fit in a matchbox. A large cluster should fit in a breadbox. Not a very scientific poll, I realize, but I'd like some numbers to use in my AAAS talk on Saturday. FWIW, this is a back-of-the-envelope calculation I did in August, when MoGo beat Myungwan Kim 8p at H9: After the match, one of the MoGo programmers mentioned that doubling the computation led to a 63% win rate against the baseline version, and that so far this scaling seemed to continue as computation power increased. So -- quick back-of-the-envelope calculation, tell me where I am wrong. 63% win rate = about half a stone advantage in go. So we need 4x processing power to increase by a stone. At the current rate of Moore's law, that's about 4 years. Kim estimated that the game with MoGo would be hard at 8 stones. That suggests that in 32 years a supercomputer comparable to the one that played in this match would be as strong as Kim. This calculation is optimistic in assuming that you can meaningfully scale the 63% win rate indefinitely, especially when measuring strength against other opponents, and not a weaker version of itself. It's also pessimistic in assuming there will be no improvement in the Monte Carlo technique. But still, 32 years seems like a surprisingly long time, much longer than the 10 years that seems intuitively reasonable. Naively, it would seem that improvements in the Monte Carlo algorithms could gain some small number of stones in strength for fixed computation, but that would just shrink the 32 years by maybe a decade. Thanks, Bob Hearn - Robert A. Hearn Neukom Institute for Computational Science, Dartmouth College robert.a.he...@dartmouth.edu http://www.dartmouth.edu/~rah/ ___ 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] Presentation of my personnal project : evolution of an artificial go player through random mutation and natural selection
Hello, I would like to share my project with you : I have developped a program trying to mimic evolution through the competition of artificial go players. The players are made of totally mutable artificial neural networks, and the compete against each other in a never ending tournament, randomly mutating and reproducing when they are successful. I have also implemented a way to share innovation among every program. I am currently looking for additional volunteer (we are 4 at the moment) to try this out. If you are interested, pleas feel free to answer here, or directly email me. I have just created a blog whose purpose will be to explain how my program work and to tell how it is going. (As for now, it has been running consistently for about a month, the players are still rather passive, trying to play patterns assuring them the greatest territory possible.) Here is the url of my blog : http://goia-hephaestos.blogspot.com/ Ernest Galbrun ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] Presentation of my personnal project : evolution of an artificial go player through random mutation and natural selection
How do you perform the neuro-evolution? What sort of genetic operators do you have? Do you have any sort of crossover? How do you represent the board and moves to the networks? - George On Fri, Feb 13, 2009 at 2:42 PM, Ernest Galbrun ernest.galb...@gmail.com wrote: Hello, I would like to share my project with you : I have developped a program trying to mimic evolution through the competition of artificial go players. The players are made of totally mutable artificial neural networks, and the compete against each other in a never ending tournament, randomly mutating and reproducing when they are successful. I have also implemented a way to share innovation among every program. I am currently looking for additional volunteer (we are 4 at the moment) to try this out. If you are interested, pleas feel free to answer here, or directly email me. I have just created a blog whose purpose will be to explain how my program work and to tell how it is going. (As for now, it has been running consistently for about a month, the players are still rather passive, trying to play patterns assuring them the greatest territory possible.) Here is the url of my blog : http://goia-hephaestos.blogspot.com/ Ernest Galbrun ___ 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] Presentation of my personnal project : evolution of an artificial go player through random mutation and natural selection
How do you perform the neuro-evolution? What sort of genetic operators do you have? Do you have any sort of crossover? How do you represent the board and moves to the networks? - George - The evolution consists in the random mutation of each neurons : weight, type of neurone, threshold, input and output adress ; the mutation probability of each propertie can mutate as well, so that the individual can eventually lock any important function. - What do you mean by genetic operator ? - Crossover is achieved through sexual reproduction. This method is always tried first, and can only occur between individual belonging to the same species. When two individual reproduce, they share randomly their genes, each gene being defined as a set of neurons. If tsis leads to some network error, the method is left and the two players are tagged as belonging to different species. - The game is fully handled by the opengo library : http://www.inventivity.com/OpenGo/ Concerning my players, the board is represented by a 19x19 integer array, each input going to one or more neuron input ; each case is also linked to one neuron output, deciding the move on any given moment. Ernest. ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] Presentation of my personnal project : evolution of an artificial go player through random mutation and natural selection
Just curious, did you ever read 'On Intelligence' by Jeff Hawkins? After reading that I got rather sold on the idea that if you're ever going to attempt making a program with neural nets that behaves intelligently then it needs to have a lot of feed-back links. Not just the standard feed-forward type of networks. Some other good ideas in that book too IMO. Mark On Feb 13, 2009, at 5:42 PM, Ernest Galbrun wrote: Hello, I would like to share my project with you : I have developped a program trying to mimic evolution through the competition of artificial go players. The players are made of totally mutable artificial neural networks, and the compete against each other in a never ending tournament, randomly mutating and reproducing when they are successful. I have also implemented a way to share innovation among every program. I am currently looking for additional volunteer (we are 4 at the moment) to try this out. If you are interested, pleas feel free to answer here, or directly email me. I have just created a blog whose purpose will be to explain how my program work and to tell how it is going. (As for now, it has been running consistently for about a month, the players are still rather passive, trying to play patterns assuring them the greatest territory possible.) Here is the url of my blog : http://goia-hephaestos.blogspot.com/ Ernest Galbrun ___ 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] Presentation of my personnal project : evolution of an artificial go player through random mutation and natural selection
On Fri, Feb 13, 2009 at 22:42, Mark Boon tesujisoftw...@gmail.com wrote: Just curious, did you ever read 'On Intelligence' by Jeff Hawkins? After reading that I got rather sold on the idea that if you're ever going to attempt making a program with neural nets that behaves intelligently then it needs to have a lot of feed-back links. Not just the standard feed-forward type of networks. Some other good ideas in that book too IMO. Mark Oh, thank you for the advice, this is the kind of things that can very smoothly be implemented in the program, I will surely a/ buy and read this book and b/ introduce some feedback interaction in my neuronal network. I have not introduced it so far because it seemed some ineffective expense in calculation power. ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
