Re: [computer-go] Odd results on 19x19
On Jan 6, 2008 11:37 PM, Don Dailey [EMAIL PROTECTED] wrote: I'm not sure I get the whole picture regarding multi-dimensional ratings. How can you compare two players with a 2-dimensional rating? You can't, so how would one use this rating? In my book, a rating's goal is to make things comparable... A 2-dimensional or more rating would be used to predict the winner. You would be able to say that a given player will beat another specified player some percentage of the time. With more than one dimension perhaps the formula would be a better predictor since it could take playing styles into consideration. It could not be used to rank players in the sense of putting them on a scale such as CGOS uses. Since there is an inherently intransitive relationship, you cannot rank players in strict order with more than 1 dimension. Thanks all who answered! I see the point, the confusion in my mind was caused by the use of the word ranking which for me does imply an ordering. best regards, Vlad ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
[computer-go] Odd results on 19x19
The styles of CS (CS-9-17-10k-1CPU), MFGO (mfgo12exp-15), and GNUGO (gnugo3.7.10_10) are different, and it's generating some odd results. Many Faces beats GnuGo 70%. There are not many games, but this is consistent with over 100 test games I've run. CS beats GnuGo 55%. Over 100 games played. CS beats Many Faces 90%. Only 20 games, but consistent with earlier results. If we look at results against GnuGo, Many Faces seems stronger than CS, but in games against CS, Many Faces is much weaker. Many Faces plays a fighting style, and CS plays a territorial style, but I'm still surprised at the difference. David ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] Odd results on 19x19
did you optimize parameters in MFGO by playing against gnugo? that'd do it. s. - Original Message From: David Fotland [EMAIL PROTECTED] To: computer-go computer-go@computer-go.org Sent: Sunday, January 6, 2008 12:52:10 PM Subject: [computer-go] Odd results on 19x19 The styles of CS (CS-9-17-10k-1CPU), MFGO (mfgo12exp-15), and GNUGO (gnugo3.7.10_10) are different, and it's generating some odd results. Many Faces beats GnuGo 70%. There are not many games, but this is consistent with over 100 test games I've run. CS beats GnuGo 55%. Over 100 games played. CS beats Many Faces 90%. Only 20 games, but consistent with earlier results. If we look at results against GnuGo, Many Faces seems stronger than CS, but in games against CS, Many Faces is much weaker. Many Faces plays a fighting style, and CS plays a territorial style, but I'm still surprised at the difference. David ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/ Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now. http://mobile.yahoo.com/;_ylt=Ahu06i62sR8HDtDypao8Wcj9tAcJ ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] Odd results on 19x19
David Fotland wrote: The styles of CS (CS-9-17-10k-1CPU), MFGO (mfgo12exp-15), and GNUGO (gnugo3.7.10_10) are different, and it's generating some odd results. Many Faces beats GnuGo 70%. There are not many games, but this is consistent with over 100 test games I've run. CS beats GnuGo 55%. Over 100 games played. CS beats Many Faces 90%. Only 20 games, but consistent with earlier results. If we look at results against GnuGo, Many Faces seems stronger than CS, but in games against CS, Many Faces is much weaker. Many Faces plays a fighting style, and CS plays a territorial style, but I'm still surprised at the difference. David ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/ I noticed that too. My feeling is that is because MF is a classical program with a global search, GNU a classical program with no global search, and Crazy Stone a MC program. MF beats GNU thanks to global search. But MF's strength without the global search (whatever that would mean) is inferior to that of GNU. CS also has a global search, so MF's global-search advantage does not work against CS. I guess that KCC Igo had the same problem as MF against Crazy Stone. I thought about a model for multi-dimensional Elo ratings once (don't give only one value to each player, but two or three, with an appropriate formula for predicting game outcome). Maybe I'll try it on CGOS data when I have time. This would not rate players along a one-dimensional line. Here is a reference to a similar idea: http://dx.doi.org/10.1016/j.jspi.2004.05.008 Abstract The Bradley–Terry model is widely and often beneficially used to rank objects from paired comparisons. The underlying assumption that makes ranking possible is the existence of a latent linear scale of merit or equivalently of a kind of transitiveness of the preference. However, in some situations such as sensory comparisons of products, this assumption can be unrealistic. In these contexts, although the Bradley–Terry model appears to be significantly interesting, the linear ranking does not make sense. Our aim is to propose a 2-dimensional extension of the Bradley–Terry model that accounts for interactions between the compared objects. From a methodological point of view, this proposition can be seen as a multidimensional scaling approach in the context of a logistic model for binomial data. Maximum likelihood is investigated and asymptotic properties are derived in order to construct confidence ellipses on the diagram of the 2-dimensional scores. It is shown by an illustrative example based on real sensory data on how to use the 2-dimensional model to inspect the lack-of-fit of the Bradley–Terry model. Rémi ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] Odd results on 19x19
steve uurtamo wrote: did you optimize parameters in MFGO by playing against gnugo? that'd do it. s. Well, I don't know about David, but I do _all_ my testing and optimizing against GNU. Rémi ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] Odd results on 19x19
My guess is that this is a combination of some intransitivity and low sample size. 100 games isn't very much data in the CS vs MFGO. As far as intransivity, perhaps Crazy Stone has some particular strength that works very well against a weakness in MFGO. The values do not make a great deal of sense, but there are a lot of unknown parameters too, such as which levels are being played by each program.Perhaps we are not comparing apples to apples? - Don David Fotland wrote: The styles of CS (CS-9-17-10k-1CPU), MFGO (mfgo12exp-15), and GNUGO (gnugo3.7.10_10) are different, and it's generating some odd results. Many Faces beats GnuGo 70%. There are not many games, but this is consistent with over 100 test games I've run. CS beats GnuGo 55%. Over 100 games played. CS beats Many Faces 90%. Only 20 games, but consistent with earlier results. If we look at results against GnuGo, Many Faces seems stronger than CS, but in games against CS, Many Faces is much weaker. Many Faces plays a fighting style, and CS plays a territorial style, but I'm still surprised at the difference. David ___ 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] Odd results on 19x19
Rémi, The idea of a non one dimension rating model is interesting. If you decide to pursue this I can give you the CGOS data in a compact format, 1 line per result. I thought of this idea too, but I didn't try to produce a model.It would be easier to test and build such a model however if you synthesized them artificially. You could purposely build a rocks/scissors/paper style into a dozen different players or more. Randomly give them different strength (using straightforward ELO ratings) but also give them one of 3 playing styles (rock, scissors, paper) in which their actual performance against a given opponent was bumped up or down 100 ELO or so depending on whether they had conflicting styles. So a paper might still beat a scissors, but it would be more difficult than their base elo ratings would suggest. Then you can play hundreds of thousands of simulated games in just seconds and generate data and see if your model can predict the results reliably. Another approach I thought of is to take a very simple game (such as tic-tac-toe) and create many players that play by simple rules but where significant transitivities might exist.You would not want the rules to be deterministic or the games would all play the same, but the rules could be probabilistic. It would be remarkable if you could capture strength characteristics with just 2 or 3 numbers instead of one. I would guess that 2 numbers might be far more accurate than 1, but with quickly diminishing returns for additional parameters. Of course it might require a huge amount of data in order to zero in on a players characteristics statistically. - Don Rémi Coulom wrote: David Fotland wrote: The styles of CS (CS-9-17-10k-1CPU), MFGO (mfgo12exp-15), and GNUGO (gnugo3.7.10_10) are different, and it's generating some odd results. Many Faces beats GnuGo 70%. There are not many games, but this is consistent with over 100 test games I've run. CS beats GnuGo 55%. Over 100 games played. CS beats Many Faces 90%. Only 20 games, but consistent with earlier results. If we look at results against GnuGo, Many Faces seems stronger than CS, but in games against CS, Many Faces is much weaker. Many Faces plays a fighting style, and CS plays a territorial style, but I'm still surprised at the difference. David ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/ I noticed that too. My feeling is that is because MF is a classical program with a global search, GNU a classical program with no global search, and Crazy Stone a MC program. MF beats GNU thanks to global search. But MF's strength without the global search (whatever that would mean) is inferior to that of GNU. CS also has a global search, so MF's global-search advantage does not work against CS. I guess that KCC Igo had the same problem as MF against Crazy Stone. I thought about a model for multi-dimensional Elo ratings once (don't give only one value to each player, but two or three, with an appropriate formula for predicting game outcome). Maybe I'll try it on CGOS data when I have time. This would not rate players along a one-dimensional line. Here is a reference to a similar idea: http://dx.doi.org/10.1016/j.jspi.2004.05.008 Abstract The Bradley–Terry model is widely and often beneficially used to rank objects from paired comparisons. The underlying assumption that makes ranking possible is the existence of a latent linear scale of merit or equivalently of a kind of transitiveness of the preference. However, in some situations such as sensory comparisons of products, this assumption can be unrealistic. In these contexts, although the Bradley–Terry model appears to be significantly interesting, the linear ranking does not make sense. Our aim is to propose a 2-dimensional extension of the Bradley–Terry model that accounts for interactions between the compared objects. From a methodological point of view, this proposition can be seen as a multidimensional scaling approach in the context of a logistic model for binomial data. Maximum likelihood is investigated and asymptotic properties are derived in order to construct confidence ellipses on the diagram of the 2-dimensional scores. It is shown by an illustrative example based on real sensory data on how to use the 2-dimensional model to inspect the lack-of-fit of the Bradley–Terry model. Rémi ___ 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] Odd results on 19x19
On Jan 6, 2008 11:00 PM, Don Dailey [EMAIL PROTECTED] wrote: The idea of a non one dimension rating model is interesting. If you decide to pursue this I can give you the CGOS data in a compact format, 1 line per result. Hi all, I'm not sure I get the whole picture regarding multi-dimensional ratings. How can you compare two players with a 2-dimensional rating? You can't, so how would one use this rating? In my book, a rating's goal is to make things comparable... best regards, Vlad ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] Odd results on 19x19
you can use a multi-d ranking system to predict the outcome of a contest between two players. this is good for handicapping, for instance. this will not necessarily create a linear ordering of the players, as you've mentioned, but it is still quite useful, and radically more efficient and useful than storing the n^2-n per-pair results. s. - Original Message From: Vlad Dumitrescu [EMAIL PROTECTED] To: computer-go computer-go@computer-go.org Sent: Sunday, January 6, 2008 5:12:56 PM Subject: Re: [computer-go] Odd results on 19x19 On Jan 6, 2008 11:00 PM, Don Dailey [EMAIL PROTECTED] wrote: The idea of a non one dimension rating model is interesting. If you decide to pursue this I can give you the CGOS data in a compact format, 1 line per result. Hi all, I'm not sure I get the whole picture regarding multi-dimensional ratings. How can you compare two players with a 2-dimensional rating? You can't, so how would one use this rating? In my book, a rating's goal is to make things comparable... best regards, Vlad ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/ Never miss a thing. Make Yahoo your home page. http://www.yahoo.com/r/hs ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] Odd results on 19x19
Vlad Dumitrescu wrote: On Jan 6, 2008 11:00 PM, Don Dailey [EMAIL PROTECTED] wrote: The idea of a non one dimension rating model is interesting. If you decide to pursue this I can give you the CGOS data in a compact format, 1 line per result. Hi all, I'm not sure I get the whole picture regarding multi-dimensional ratings. How can you compare two players with a 2-dimensional rating? You can't, so how would one use this rating? In my book, a rating's goal is to make things comparable... best regards, Vlad The idea is that players would not be ranked on a linear scale, but we would have a formula to estimate the probability of winning between any pair of players. For instance, if player A has rating (A1, A2, A3) and player B has rating (B1, B2, B3) Delta = ((A1-B1)^3 + (A2-B2)^3 + (A3-B3)^3) / ((A1-B1)² + (A2-B2)² + (A3-B3)²) P(A beats B) = 1 / (1 + exp(-Delta)) if A1 = A2 = A3 and B1 = B2 = B3, this is the usual Bradley-Terry model. But with 3 dimensions, it is possible to get a cycle for instance with: A=(1, -1, 0) B=(-1, 0, 1) C=(0, 1, -1) With these ratings and the formula above, P(A beats B) 0.5, P(B beats C) 0.5, and P(C beats A) 0.5. It is exactly the same principle as the basic Bradley-Terry model. The very big difficulty is finding the maximum-a-posteriori of the ratings from the observation of game results. There is no easy optimization algorithm like for the one-dimensional model. The probability distribution has many local optima, so it is tricky. Rémi ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] Odd results on 19x19
Vlad Dumitrescu wrote: On Jan 6, 2008 11:00 PM, Don Dailey [EMAIL PROTECTED] wrote: The idea of a non one dimension rating model is interesting. If you decide to pursue this I can give you the CGOS data in a compact format, 1 line per result. Hi all, I'm not sure I get the whole picture regarding multi-dimensional ratings. How can you compare two players with a 2-dimensional rating? You can't, so how would one use this rating? In my book, a rating's goal is to make things comparable... A 2-dimensional or more rating would be used to predict the winner. You would be able to say that a given player will beat another specified player some percentage of the time. With more than one dimension perhaps the formula would be a better predictor since it could take playing styles into consideration. It could not be used to rank players in the sense of putting them on a scale such as CGOS uses. Since there is an inherently intransitive relationship, you cannot rank players in strict order with more than 1 dimension. - Don best regards, Vlad ___ 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] Odd results on 19x19
I've mentioned this before, but hopefully not recently enough to make this annoying. Computer go people and corewars people overlap somewhat. Intransitivity is extremely important for corewars, making?corewars a good domain to study it. Here is an example of a nice graphical way to visualize intransitivities between corewars programs. http://www.koth.org/lcgi-bin/hugetable.pl?hill94x? In corewars, you might look at a table like this and say Oh, I'm losing too many games against x-type strategy, I need to concentrate on that aspect of my bot. Or you might say, I'm doing great against y-type strategy, and I'm betting there will be a lot of those in the upcoming tournament. It's more than predicting the outcome of a particular matchup, or a ranking in a static field. We haven't seen strong statistical evidence for intransitivity in computer go, but I don't think anyone has looked very hard yet. - Dave Hillis -Original Message- From: Vlad Dumitrescu [EMAIL PROTECTED] To: computer-go computer-go@computer-go.org Sent: Sun, 6 Jan 2008 5:12 pm Subject: Re: [computer-go] Odd results on 19x19 On Jan 6, 2008 11:00 PM, Don Dailey [EMAIL PROTECTED] wrote: The idea of a non one dimension rating model is interesting. If you decide to pursue this I can give you the CGOS data in a compact format, 1 line per result. Hi all, I'm not sure I get the whole picture regarding multi-dimensional ratings. How can you compare two players with a 2-dimensional rating? You can't, so how would one use this rating? In my book, a rating's goal is to make things comparable... best regards, Vlad ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/ More new features than ever. Check out the new AIM(R) Mail ! - http://webmail.aim.com ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] Odd results on 19x19
We haven't seen strong statistical evidence for intransitivity in computer go, but I don't think anyone has looked very hard yet. It seems like it probably exists to some degree - It would be interesting to study this. - Don [EMAIL PROTECTED] wrote: I've mentioned this before, but hopefully not recently enough to make this annoying. Computer go people and corewars people overlap somewhat. Intransitivity is extremely important for corewars, making corewars a good domain to study it. Here is an example of a nice graphical way to visualize intransitivities between corewars programs. http://www.koth.org/lcgi-bin/hugetable.pl?hill94x In corewars, you might look at a table like this and say Oh, I'm losing too many games against x-type strategy, I need to concentrate on that aspect of my bot. Or you might say, I'm doing great against y-type strategy, and I'm betting there will be a lot of those in the upcoming tournament. It's more than predicting the outcome of a particular matchup, or a ranking in a static field. We haven't seen strong statistical evidence for intransitivity in computer go, but I don't think anyone has looked very hard yet. - Dave Hillis -Original Message- From: Vlad Dumitrescu [EMAIL PROTECTED] To: computer-go computer-go@computer-go.org Sent: Sun, 6 Jan 2008 5:12 pm Subject: Re: [computer-go] Odd results on 19x19 On Jan 6, 2008 11:00 PM, Don Dailey [EMAIL PROTECTED] mailto:[EMAIL PROTECTED] wrote: The idea of a non one dimension rating model is interesting. If you decide to pursue this I can give you the CGOS data in a compact format, 1 line per result. Hi all, I'm not sure I get the whole picture regarding multi-dimensional ratings. How can you compare two players with a 2-dimensional rating? You can't, so how would one use this rating? In my book, a rating's goal is to make things comparable... best regards, Vlad ___ computer-go mailing list computer-go@computer-go.org mailto:computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/ More new features than ever. Check out the new AIM(R) Mail http://o.aolcdn.com/cdn.webmail.aol.com/mailtour/aol/en-us/text.htm?ncid=aimcmp000501! ___ 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/