Re: [computer-go] Technical Report on MoGo
Hello, I'd be a bit more careful about the comparison with alpha-beta in section 2.3. I believe that iterative deepening of alpha-beta is very common. It can be argued that when iterative deepening is used, an early termination isn't very detrimental. [...] Alpha-Beta is for practical reasons of course also an anytime algorithm. [...] .My reaction when I read this statement was: iterative deepening is not yet invented in the Go community. Of course iterative deepening exists. But to me it does not make Alpha-Beta algorithm an anytime algorithm. First because the unit (one iteration) costs much more in alpha-beta. By iteration I mean that if you stop your program during an iteration, then it behaves as in the last iteration (the current iteration is lost). In MC/UCT, the iteration takes less than 1 ms. Second, and more importantly, the time increase of the iteration is huge in alpha-beta. The time to perform the search at depth k+1 is much higher than for depth k. So for me the reasons we gave comparing to alpha-beta hold, even if you are right by saying that we should have mentionned iterative deepening. Sylvain ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
RE: [computer-go] Technical Report on MoGo
I'd be a bit more careful about the comparison with alpha-beta in section 2.3. I believe that iterative deepening of alpha-beta is very common. It can be argued that when iterative deepening is used, an early termination isn't very detrimental. I've seen people get completely turned off to a paper simply because they compare their carefully optimized results to a poor implementation of some other algorithm (ie. alpha beta). -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED] Sent: Friday, December 01, 2006 6:03 PM To: computer-go@computer-go.org Subject: [computer-go] Technical Report on MoGo Hello all, as perhaps some of you may be interested, I give here a link to a technical report about MoGo. You can find there a lot of details about the ideas around MoGo. While we tried to be as clear as possible, some details may lack. There is still no number on this report, but this will come in a few days. http://hal.inria.fr/inria-00117266 I would like to thank of course all the authors, but also Rémi Coulom who shared a lot of details about CrazyStone and his ideas. I also would like to thank all the contributors in this list for interesting discussions. Now my feeling is that the improving random simulations part of this work is promising. We have only done very few steps in this direction, and it gives quite convincing results. It was what I meant in the random distribution discussions we have in this list. I am pretty sure that making improvements in this direction would increase a lot the level of MC players even (or especially) in 19x19. And this can be done very soon (well, perhaps not before sunday :)). Sylvain ___ 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] Technical Report on MoGo
- Original Message - From: House, Jason J. [EMAIL PROTECTED] To: [EMAIL PROTECTED]; computer-go computer-go@computer-go.org Sent: Monday, December 04, 2006 8:10 PM Subject: RE: [computer-go] Technical Report on MoGo I'd be a bit more careful about the comparison with alpha-beta in section 2.3. I believe that iterative deepening of alpha-beta is very common. It can be argued that when iterative deepening is used, an early termination isn't very detrimental. I've seen people get completely turned off to a paper simply because they compare their carefully optimized results to a poor implementation of some other algorithm (ie. alpha beta). Alpha-Beta is for practical reasons of course also an anytime algorithm. In chess one does not send the first iterations to the GUI, because showing the result in the GUI is slower than the calculation of the engine. And the user would not notice it anyway, its too fast. My reaction when I read this statement was: iterative deepening is not yet invented in the Go community. But Alpha-Beta is not a continous algorithm. If one searches to depth k, the nodes till the first move at depth k+1 is completly searched have no additional information. Usually one does some estimates beforehand if it pays to search for another iteration. Chrilly -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of [EMAIL PROTECTED] Sent: Friday, December 01, 2006 6:03 PM To: computer-go@computer-go.org Subject: [computer-go] Technical Report on MoGo Hello all, as perhaps some of you may be interested, I give here a link to a technical report about MoGo. You can find there a lot of details about the ideas around MoGo. While we tried to be as clear as possible, some details may lack. There is still no number on this report, but this will come in a few days. http://hal.inria.fr/inria-00117266 I would like to thank of course all the authors, but also Rémi Coulom who shared a lot of details about CrazyStone and his ideas. I also would like to thank all the contributors in this list for interesting discussions. Now my feeling is that the improving random simulations part of this work is promising. We have only done very few steps in this direction, and it gives quite convincing results. It was what I meant in the random distribution discussions we have in this list. I am pretty sure that making improvements in this direction would increase a lot the level of MC players even (or especially) in 19x19. And this can be done very soon (well, perhaps not before sunday :)). Sylvain ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/ ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/ ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
Re: [computer-go] Technical Report on MoGo
Quoting [EMAIL PROTECTED]: Now my feeling is that the improving random simulations part of this work is promising. We have only done very few steps in this direction, and it gives quite convincing results. It was what I meant in the random distribution discussions we have in this list. I am pretty sure that making improvements in this direction would increase a lot the level of MC players even (or especially) in 19x19. And this can be done very soon (well, perhaps not before sunday :)). I just read through your paper eagerly, and found that the your changes to the random simulations are pretty much exactly the same in principle as most things Valkyria does. The difference is that my hardcoded patterns are fewer simply because I have not implemeted all yet. Valkyria also checks stones freshly in atari for good moves to save them and tests hardcoded patterns directly near the last move. The reason this is important for MC is that otherwise move sequences that never occur in real game occur frequently such as cutting a diagonal connection without a double threat. Without such knowledge about basic patterns MC program tends to play too strong shapes. I am now currently working with a system for matching larger but netherthelees very fast patterns, which I might be able to use in the random simualations as well as in the UCT-tree. The things that were different from Valkyria was a little difficult to get the first time but I will read it through more carefully soon and come back with more comments. -Magnus ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/
[computer-go] Technical Report on MoGo
Hello all, as perhaps some of you may be interested, I give here a link to a technical report about MoGo. You can find there a lot of details about the ideas around MoGo. While we tried to be as clear as possible, some details may lack. There is still no number on this report, but this will come in a few days. http://hal.inria.fr/inria-00117266 I would like to thank of course all the authors, but also Rémi Coulom who shared a lot of details about CrazyStone and his ideas. I also would like to thank all the contributors in this list for interesting discussions. Now my feeling is that the improving random simulations part of this work is promising. We have only done very few steps in this direction, and it gives quite convincing results. It was what I meant in the random distribution discussions we have in this list. I am pretty sure that making improvements in this direction would increase a lot the level of MC players even (or especially) in 19x19. And this can be done very soon (well, perhaps not before sunday :)). Sylvain ___ computer-go mailing list computer-go@computer-go.org http://www.computer-go.org/mailman/listinfo/computer-go/