I'm an AGA 5d, played about 20 9x9 games with MoGo at kgs recently. It is not easy to win, not easy to bully MoGo. My experience: 1. play white much easy to win, 7.5 is a big komi. 2. 10 minutes sometimes a little short to me.
I think play 10 minutes games against AGA 6d seriously, MoGo have a chance to win 1 or 2. Maybe MoGo play black and no komi is a nice match. Peter817 -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Don Dailey Sent: Tuesday, April 03, 2007 9:17 PM To: computer-go Subject: Re: Re:[computer-go] MoGo This article didn't specify the boardsize. This has no chance of being interesting unless it's played on a 9x9 board. If it is on a 19x19 board, it's pretty much a silly exhibition match that has no point. If it's on a 9x9 board, it starts to be more interesting, but assuming that is the case, the experiment could be much more interesting. Notice the disparity in the two claims: 1. A claim that it's "near the level of a professional" 2. A counter-claim it isn't even close to 1 dan One or both of these claims is way off base. If Mogo is humiliated in this match, to the average person it will appear as proof of the second claim. The average person is far too impressed by poorly constructed experiments like this. Of course I believe Mogo is easily 1 dan, perhaps much stronger at 9x9. But I would find a 9x9 SERIOUS match against a 1 or 2 dan player far more interesting. That's because I think Mogo would win such a match. Then we could say, "there is strong evidence to indicate Mogo is playing at least in the 2 dan range or higher!" On the other hand, if Mogo loses against a 6 dan player, there is not much that can be said (unless of course the match is close.) But I know from personal experience that close matches are never taken seriously. It will get "spun" as the human being taken by surprise against a much weaker but spunky opponent, and then when the human decided to take the match seriously, it was no contest. If the computer finishes strong but loses, the spin is that the human, after winning the first few games easily, let his guard down (or decided to have fun with his opponent by playing speculated experimental moves to "see what the computer would do.") I'm excited of course that there is a serious match - at least Mogo is strong enough to spark this kind of interest. Of course I would love to see Mogo win, but as optimistic as I am of computer progress with MC and UCT I don't see that there is much of a chance against a 6 dan. I hope I am wrong! On the other hand, Mogo is so strong, that it can't be reasonably tested on CGOS. Weak version of Mogo that play almost instantly are being tested on CGOS and there is large strength gap between them and the next best programs. So it would be fun to have a realistic test of how strong Mogo is. Perhaps it IS close to a 6 dan in strength, presumably they have made many improvements and continue to make them. If Mogo is strong enough to win even 1 game, I think it will be at least a somewhat interesting result. - Don On Tue, 2007-04-03 at 15:59 -0700, terry mcintyre wrote: > Sylvain, > > Were you aware of this challenge from the American Go Association? The > following is from the latest AGA newsletter; you can send corrections > or replies to [EMAIL PROTECTED] > > GO ONLINE: MoGo -- No-Go, So-So or Uh-Oh? > Go has been called "The fruit fly of IT", and for a good reason -- > although software engineers have created programs that can defeat the > strongest chess players, the strongest go programs are routinely > defeated by talented children. In fact, go is the lone holdout, the > only classic game that has not yet been "solved" (so to speak) for the > computer. If you wonder why, the Wikipedia article on computer go is a > good place to start. > One way to simplify the problem is to work with a smaller board, > an approach followed by Levente Kocsis and Csaba Szepesvari, who are > working together at the Hungarian Academy of Science on a 9x9 program > called MoGo. Their recent claim that MoGo plays 9x9 go "near the level > of a professional go player made international news so we decided to > investigate. > Sylvain Gelly, a contributor to the program, clarified the "one-armed > bandit strategy," a variation of the ancient Chinese proverb "Rich men > don't pick fights." Gelly told the EJ that "MoGo tries to maximize its > winning probability. When behind, MoGo will play 'strange' moves to > try to catch up, and when ahead, it will prefer safe moves which > secure victory instead of keeping score. Usually it loses points when > ahead, trading profit for safety, aiming to win by +0.5." To learn > more see the Sensei's Library MoGo page. > MoGo has played extensively on the Internet but evidence that it > plays beyond the mid-kyu level is not compelling, so we're going to > put MoGo to the test. Philip Waldron -- a solid 6-dan with a current > AGA rating of 6.47 who has reviewed go software for the EJ -- will > play a best-of-seven series against MoGo in the computer go room on > KGS. Game times will not be announced in advance, and times will vary > to eliminate the possibility of human interference on the MoGo side. > The results, and possibly a few of the games, will appear in a future > EJ, so stay tuned! > - Roy Laird > > > > > > ______________________________________________________________________ > It's here! Your new message! > Get new email alerts with the free Yahoo! Toolbar. > _______________________________________________ > 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/
