EPT![[image loading]](http://i.imgur.com/amCX3vz.jpg)
For anyone who was curious, the message that AlphaGo displays on the screen when it resigns is, "AlphaGo resigns" and "The result 'W+resign' was added to the game information."
Go Igo Weiqi Baduk - Page 20
| Forum Index > General Games |
|
usopsama
6502 Posts
For anyone who was curious, the message that AlphaGo displays on the screen when it resigns is, "AlphaGo resigns" and "The result 'W+resign' was added to the game information." | ||
|
Yurie
11880 Posts
On March 14 2016 04:29 usopsama wrote: For anyone who was curious, the message that AlphaGo displays on the screen when it resigns is, "AlphaGo resigns" and "The result 'Weresign' was added to the game information." Minor correction: "The result 'W + Resign' was added to the game information." For white win by black resigning. | ||
The_Red_Viper
19533 Posts
 | ||
|
JeffKim
Korea (South)36 Posts
On March 14 2016 06:09 The_Red_Viper wrote: Once. It won other times.alphago lost? interesting | ||
|
The_Red_Viper
19533 Posts
On March 14 2016 07:54 JeffKim wrote: Once. It won other times. Yes i know, i just thought this might be a 5:0 | ||
Fecalfeast
Canada11355 Posts
| ||
|
evilfatsh1t
Australia8690 Posts
successive wins from lee sedol could mean he noticed some patterns within the program | ||
|
sc14s
United States5052 Posts
On March 14 2016 11:51 evilfatsh1t wrote: if lee sedol wins the next game i would think that if you turned this into a best of 7 lee sedol would comeback to win 3-4. successive wins from lee sedol could mean he noticed some patterns within the program All that would mean is AlphaGo would go back into its Hyperbolic time chamber for a little longer to improve some more.  | ||
|
emperorchampion
Canada9496 Posts
| ||
|
strongwind
United States862 Posts
I'm glad he got one. No question that going up against 1200 CPUs is no small feat. Hopefully he can pull out some more God moves in the next one. | ||
|
Glacierz
United States1244 Posts
http://www.goratings.org/ | ||
|
The_Red_Viper
19533 Posts
On March 14 2016 23:43 Glacierz wrote: Alphago is now on the ranking board: http://www.goratings.org/ Was Sedol first before he lost to a "newcomer"? Not that Elo matters for the pros (i would assume) but that would suck nonetheless ^^ | ||
|
mahrgell
Germany3943 Posts
On March 15 2016 01:01 The_Red_Viper wrote: Was Sedol first before he lost to a "newcomer"? Not that Elo matters for the pros (i would assume) but that would suck nonetheless ^^ No, he wasn't. But this list is nothing official anyway (but made by a french programmer) and I wouldn't bet on Lee Sedol even knowing it  | ||
|
Frolossus
United States4779 Posts
On March 15 2016 01:01 The_Red_Viper wrote: Was Sedol first before he lost to a "newcomer"? Not that Elo matters for the pros (i would assume) but that would suck nonetheless ^^ sedol was rank 3 before this | ||
|
Torte de Lini
Germany38463 Posts
On March 14 2016 23:43 Glacierz wrote: Alphago is now on the ranking board: http://www.goratings.org/ haha oh god | ||
|
meatpudding
Australia520 Posts
On March 14 2016 11:51 evilfatsh1t wrote: if lee sedol wins the next game i would think that if you turned this into a best of 7 lee sedol would comeback to win 3-4. successive wins from lee sedol could mean he noticed some patterns within the program I think it's likely for Lee Sedol to win game 5 as well. The match is inherently imbalanced as the commentators once said. Alphago has been trained on existing matches, including those from Lee Sedol. But this version of Alphago has never been shown before. | ||
|
Torte de Lini
Germany38463 Posts
| ||
|
Excalibur_Z
United States12237 Posts
On March 15 2016 09:10 meatpudding wrote: I think it's likely for Lee Sedol to win game 5 as well. The match is inherently imbalanced as the commentators once said. Alphago has been trained on existing matches, including those from Lee Sedol. But this version of Alphago has never been shown before. They mentioned in the press conference that all of the games that AlphaGo used to train were amateur matches pulled from online Go sites. No professional games, by Lee Sedol or otherwise, were included in its learning database. | ||
|
meatpudding
Australia520 Posts
On March 15 2016 10:17 Excalibur_Z wrote: They mentioned in the press conference that all of the games that AlphaGo used to train were amateur matches pulled from online Go sites. No professional games, by Lee Sedol or otherwise, were included in its learning database. I was under the impression that the network was trained with ancient traditional games, online games as well as professional games. But I don't know for sure. From the abstract of the Nature article, These deep neural networks are trained by a novel combination of supervised learning from human expert games, and reinforcement learning from games of self-play. I would be interested to find out which games it has used to train on. If it really didn't use any 9-dan games then that's really impressive to reach the level that it has. | ||
|
mahrgell
Germany3943 Posts
On March 15 2016 13:12 meatpudding wrote: I was under the impression that the network was trained with ancient traditional games, online games as well as professional games. But I don't know for sure. From the abstract of the Nature article, I would be interested to find out which games it has used to train on. If it really didn't use any 9-dan games then that's really impressive to reach the level that it has. They used a very large database of KGS 4d-9d games. Can I now say that I was part of Alphagos development, because some of my games were in that database? :D | ||
| ||
