| Life In 19x19 http://www.lifein19x19.com/ |
|
| The Shodan Go Bet http://www.lifein19x19.com/viewtopic.php?f=18&t=2646 |
Page 1 of 5 |
| Author: | yoyoma [ Mon Dec 20, 2010 12:07 pm ] |
| Post subject: | The Shodan Go Bet |
In 1997, John Tromp made a bet with Darren Cook that he would not lose a serious match with a computer Go player before the end of 2010. Now 13 years later the match has been scheduled to start December 28th. Bet details: http://dcook.org/gobet/details.html Back then programs were not very good, and progress was slow. But in 2006 there was a breakthrough technique -- Monte Carlo tree search. Since then bots have improve very rapidly, about a stone every year, and Zen19 is 3d on KGS now (see http://senseis.xmp.net/?KGSBotRatings for a history of bot progress). So who do you think will win the bet? What do you think about the future of computer go? ------------------------- Update: Links to the games and result: |
|
| Author: | rubin427 [ Mon Dec 20, 2010 1:00 pm ] |
| Post subject: | Re: The Shodan Go Bet |
From the article: Quote: John (Tromp) is now 2-dan (European), going on 3-dan, understands computer go, and will certainly know how to exploit any known weaknesses of the computer. Could be a close match if the Zen engine is used. |
|
| Author: | Javaness [ Mon Dec 20, 2010 1:45 pm ] |
| Post subject: | Re: The Shodan Go Bet |
Did Cook pick a bot yet? I heard MFOG might be used, but I don't see confirmation anywhere yet |
|
| Author: | hyperpape [ Mon Dec 20, 2010 2:05 pm ] |
| Post subject: | Re: The Shodan Go Bet |
One of the linked pages says that the bot will be announced just before the competition, perhaps to prevent Tromp from doing focused training against it (though he's able to play any available bot right now). |
|
| Author: | palapiku [ Mon Dec 20, 2010 2:34 pm ] |
| Post subject: | Re: The Shodan Go Bet |
Zen19 is currently at exactly the dividing line between 3d and 4d. |
|
| Author: | Magicwand [ Mon Dec 20, 2010 4:06 pm ] |
| Post subject: | Re: The Shodan Go Bet |
i dont think computer can reach real human 3dan level for next 10 years. |
|
| Author: | palapiku [ Mon Dec 20, 2010 4:14 pm ] |
| Post subject: | Re: The Shodan Go Bet |
Magicwand wrote: i dont think computer can reach real human 3dan level for next 10 years. I'm not quite sure what you mean, since Zen19 is already 3d KGS and I don't know how 3d KGS is different from "real human 3d", but I would easily bet $1000 or even $10000 on that, if you like. (I don't want to wait 10 years, but I doubt I'd have to wait much more than a year before computers convincingly dominate any "real human 3d".) |
|
| Author: | Magicwand [ Mon Dec 20, 2010 4:32 pm ] |
| Post subject: | Re: The Shodan Go Bet |
how did computer earn kgs 3d ?? did he actually play other players and beat them? does that mean that he can play in two stone and win? if computer is 3d kgs then i should have hard time win against him right? if you like betting i will bet $100 and play him 3 stone. are you so confident that you will win? i wouldnt bet my $100 if it was actual human playin.. that is what i mean. |
|
| Author: | oren [ Mon Dec 20, 2010 4:36 pm ] |
| Post subject: | Re: The Shodan Go Bet |
http://www.gokgs.com/gameArchives.jsp?user=zen19 You can decide yourself looking at the games. |
|
| Author: | Magicwand [ Mon Dec 20, 2010 4:47 pm ] |
| Post subject: | Re: The Shodan Go Bet |
oren wrote: http://www.gokgs.com/gameArchives.jsp?user=zen19 You can decide yourself looking at the games. thank you for your info. i dont think computer can beat me in 4 stone... it is more like 1k or 2k to me. |
|
| Author: | tapir [ Mon Dec 20, 2010 4:55 pm ] |
| Post subject: | Re: The Shodan Go Bet |
zen has a suspicious number of wins by time. something like 20% (no losses by time). at what time limits are the games played, why have the human players so much trouble with the time? |
|
| Author: | Magicwand [ Mon Dec 20, 2010 5:03 pm ] |
| Post subject: | Re: The Shodan Go Bet |
tapir wrote: zen has a suspicious number of wins by time. something like 20% (no losses by time). at what time limits are the games played, why have the human players so much trouble with the time? 3min 5x15sec |
|
| Author: | palapiku [ Mon Dec 20, 2010 5:16 pm ] |
| Post subject: | Re: The Shodan Go Bet |
Magicwand wrote: if you like betting i will bet $100 and play him 3 stone. are you so confident that you will win? No  But you said ten years... I'm sure that within ten years a computer will be able to beat you in an even game at any time setting. |
|
| Author: | tapir [ Mon Dec 20, 2010 5:23 pm ] |
| Post subject: | Re: The Shodan Go Bet |
Magicwand wrote: tapir wrote: zen has a suspicious number of wins by time. something like 20% (no losses by time). at what time limits are the games played, why have the human players so much trouble with the time? 3min 5x15sec well, they say the computer gains one stone in strength by doubling the time. how much loses a human by playing in such a blitz mode? more than one i guess. i mean can they read a fight over ladder breakers at such time limit - for how much moves? add the real timeouts (as opposed to those preferring to see a lost by time over an resignation) ... still a feat, but i would be way more impressed by this rank at 15min 3x30sec or something like that. |
|
| Author: | Magicwand [ Mon Dec 20, 2010 5:24 pm ] |
| Post subject: | Re: The Shodan Go Bet |
palapiku wrote: Magicwand wrote: if you like betting i will bet $100 and play him 3 stone. are you so confident that you will win? No But you said ten years... I'm sure that within ten years a computer will be able to beat you in an even game at any time setting. yes ten years. personally i dont think computer will beat me in my life time. i am 43 years old and planning to live upto 100. i guess you have 57 more years
|
|
| Author: | yoyoma [ Mon Dec 20, 2010 5:45 pm ] |
| Post subject: | Re: The Shodan Go Bet |
Zen19b [3d] plays 0+5x:10 Zen19 [3d] plays 3+3x:15 Zen19n [2d] plays 20+5x:30 So it looks like humans benefit more than Zen19 from longer thinking times. |
|
| Author: | SpongeBob [ Mon Dec 20, 2010 5:47 pm ] |
| Post subject: | Re: The Shodan Go Bet |
Magicwand wrote: personally i dont think computer will beat me in my life time. i am 43 years old and planning to live upto 100. i guess you have 57 more years As much as I would like your words to become true - you are too optimistic here. You (and I) have no idea about the progress that information technology will bring in the next decades. THE SINGULARITY IS NEAR |
|
| Author: | quantumf [ Tue Dec 21, 2010 1:08 am ] |
| Post subject: | Re: The Shodan Go Bet |
tapir wrote: zen has a suspicious number of wins by time. something like 20% (no losses by time). at what time limits are the games played, why have the human players so much trouble with the time? I would guess bots gain at least a stone (perhaps even 2) by always playing to their ability (i.e. they don't make blunders, they don't get bored, they don't get drunk, they don't get distracted, etc). |
|
| Author: | shapenaji [ Tue Dec 21, 2010 1:55 am ] |
| Post subject: | Re: The Shodan Go Bet |
I dunno, I think the statistic "A computer gains one stone by doubling their time" is a little misleading, I don't think Zen gains that much with repeated doubling of computing time. Human players, otoh, can lose substantially more than one stone by halving their time. If you include the sampling bias that players who beat a bot are less likely to play it in future, I suspect that the Zenbot is taking advantage of "Time-fish". |
|
| Author: | quantumf [ Tue Dec 21, 2010 2:16 am ] |
| Post subject: | Re: The Shodan Go Bet |
shapenaji wrote: I dunno, I think the statistic "A computer gains one stone by doubling their time" is a little misleading, I don't think Zen gains that much with repeated doubling of computing time. Human players, otoh, can lose substantially more than one stone by halving their time. I think what the quote means is that the computer gains one stone by doubling its time, all other things being equal (in other words, the opponent doesn't get to double their time). Put another way, the program gains one stone compared to a known and stable set of fixed rank opponents, e.g. gnugo version x with thinking time y. So yes, I agree it is misleading. |
|
| Page 1 of 5 | All times are UTC - 8 hours [ DST ] |
| Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group http://www.phpbb.com/ |
|