[Computer-go] Game Over

[Rémi Coulom]

https://www.youtube.com/watch?v=g-dKXOlsf98

Google beats Fan Hui, 2 dan pro, 5-0 (19x19, no handicap)!
Congratulations! I am proud of my student Aja. They'll play Lee Sedol in
March.

It's a pity they don't participate in the UEC Cup.

I read the paper. The most original idea is in learning a value network.
It seems to be extremely efficient.

Rémi
_______________________________________________
Computer-go mailing list
Compu...@computer-go.org
http://computer-go.org/mailman/listinfo/computer-go

[Darren Cook]

> Google beats Fan Hui, 2 dan pro, 5-0 (19x19, no handicap)!

> ...
> I read the paper...

Is it available online anywhere, or only in Nature?

I just watched the video, which was very professionally done, but didn't
come with the SGFs, information on time limits, number of CPUs, etc.
Aja, David - surely the NDAs no longer apply, and you can now tell us
all the details?

Darren

P.S. Curiously the BBC ran an article today on how Facebook is getting
close to top pro level too: http://www.bbc.co.uk/news/technology-35419141

[Erik S. Steinmetz]

This seems quite amazing. Congratulations to the Google DeepMind team and AlphaGo!

Rémi, Is the paper of which you speak available?

Many thanks,

Erik S. Steinmetz
er...@steinmetz.org
(612) 789-6940
(612) 978-4342 cell

[Julian Schrittwieser]

http://www.nature.com/nature/journal/v529/n7587/full/nature16961.html

well done Aja :)

[John Tromp]

I foresee a future where we watch Google vs Facebook matches with
human professionals providing commentary on their superiors :-)

Interesting times we live in!

-John

[Gian-Carlo Pascutto]

On 27/01/2016 18:58, Darren Cook wrote:
> P.S. Curiously the BBC ran an article today on how Facebook is getting
> close to top pro level too: http://www.bbc.co.uk/news/technology-35419141

http://googleresearch.blogspot.be/2016/01/alphago-mastering-ancient-game-of-go.html

"The match was played behind closed doors between October 5-9 last year."

They already achieved this a while ago. The Google announcement looks
like a direct response to the press Facebook was getting.

--
GCP

[Hiroshi Yamashita]

Distributed AlphaGo is stronger than CrazyStone by +1200 Elo?!

AlphaGo: Mastering the ancient game of Go with Machine Learning
http://googleresearch.blogspot.jp/2016/01/alphago-mastering-ancient-game-of-go.html

Hiroshi Yamashita

[Julian Schrittwieser]

Actually the paper has been in the works for quite a while and was already set to be released today for some weeks.

It seems a journalist reached out to Facebook to comment a day ago.

[Marc Landgraf]

for those looking for sgfs: http://deepmind.com/alpha-go.html

[Julian Schrittwieser]

If you want to view them in the browser, I've also put them on my blog: http://www.furidamu.org/blog/2016/01/26/mastering-the-game-of-go-with-deep-neural-networks-and-tree-search/ (scroll down)

[Yoshiki Ohshima]

Thank you for the game records! I really am just a by stander and
kibizer and a weak player, but isn't the style of Fan Hui going too
low positions, keep making clusters, and do a local fight at a time?

--
-- Yoshiki

[Rémi Coulom]

https://storage.googleapis.com/deepmind-data/assets/papers/deepmind-mastering-go.pdf

On 01/27/2016 06:58 PM, Darren Cook wrote:
> Is it available online anywhere, or only in Nature?

_______________________________________________

[Hideki Kato]

Congratulation! Really an excellent job, David and Aja!

I imagined once but didn't think such value networks can be trained in
practice, what a suprising machine power of the cloud!

Hideki

Remi Coulom: <56A919E2...@free.fr>:

--
Hideki Kato <mailto:hideki...@ybb.ne.jp>

[Ryan Grant]

To the authors: Did the deep-NN architecture learn ladders on its own,
or was any extra ladder-evaluation code added to the playout module?

[Álvaro Begué]

It's in the paper: "ladder capture" and "ladder escape" are features that are fed as inputs into the CNN.

Álvaro.

[Yuandong Tian]

Congratulations to Aja & DeepMind team! Amazing results :)

----------------------------

Yuandong Tian

Research Scientist,

Facebook Artificial Intelligence Research (FAIR)

Website: https://research.facebook.com/researchers/1517678171821436/yuandong-tian/

[Thomas Wolf]

Congratulations to Aja $ DeepMind to that great result!

I am curious to see AlphaGo having to play a tough narrow endgame. In the
first of the 5 games it could affort not to play totally optimal in the end
and in the next 4 games Fan resigned. End games require again other, more math like
skills, at least as human player. But maybe trained networks got good at that too.

Thomas

[Anders Kierulf]

Congrats to the AlphaGo team — a tremendous accomplishment!

I've been reading the paper and have written up a summary of what they did:

https://smartgo.com/blog/google-alphago.html

Please let me know if I misinterpreted anything. Also, the truncated rollouts mentioned in the paper are still unclear to me.

Thanks,
Anders

["Ingo Althöfer"]

Hello Anders,

thanks for the summary on the smartgo site.

> ... the truncated rollouts mentioned in the paper are still unclear to me.

The greatest expert on these rollouts might be Richard Lorentz.
He applied them successfully to his bots in the games Amazons (not to be mixed up
with the online bookshop), Havannah and Breakthrough. Richard found that in many
applications a truncation level of 4 moves seem to work quite well.
There is a paper by him on this topic in the proceedings of the conference
Advances in Computer Games 2015 (in Leiden , NL), published by Springer
Lecture Notes in Computer Science (LNCS).

A very early application of truncated rollouts was applied by Brian
Sheppard in his bot for Scrabble (MAVEN).

Ingo.

["Ingo Althöfer"]

Click here for the slides of Richard's talk:

https://acg2015.files.wordpress.com/2015/07/lorentz.pdf

[Robert Jasiek]

On 28.01.2016 04:57, Anders Kierulf wrote:
> Please let me know if I misinterpreted anything.

You write "Position evaluation has not worked well for Go in the past"
but I think you should write "...Computer Go..." because applicable,
reasonably accurate theory for human players' positional evaluation
exists, see e.g. my two books Positional Judgement.

--
robert jasiek

[Darren Cook]

> If you want to view them in the browser, I've also put them on my blog:
> http://www.furidamu.org/blog/2016/01/26/mastering-the-game-of-go-with-deep-neural-networks-and-tree-search/
> (scroll down)

Thanks. Has anyone (strong) made commented versions yet? I played
through the first game, but it just looks like a game between two
players much stronger than me :-)

(Ingo, are you analyzing them with e.g. CrazyStone? Is there a
particular point where it adjusts who it thinks is winning?)

Darren

[Xavier Combelle]

here a comment by Antti Törmänen  http://gooften.net/2016/01/28/the-future-is-here-a-professional-level-go-ai/

[J. van der Steen]

Hi Xavier,

Really nice comments by Antti Törmänen, to the point and very clear
explanation. Thanks for the pointer.

best regards,
Jan van der Steen

On 28-01-16 11:45, Xavier Combelle wrote:
> here a comment by Antti Törmänen
> http://gooften.net/2016/01/28/the-future-is-here-a-professional-level-go-ai/
>
> 2016-01-28 11:19 GMT+01:00 Darren Cook <dar...@dcook.org

> <mailto:dar...@dcook.org>>:

>
> > If you want to view them in the browser, I've also put them on my blog:
> >http://www.furidamu.org/blog/2016/01/26/mastering-the-game-of-go-with-deep-neural-networks-and-tree-search/
> > (scroll down)
>
> Thanks. Has anyone (strong) made commented versions yet? I played
> through the first game, but it just looks like a game between two
> players much stronger than me :-)
>
> (Ingo, are you analyzing them with e.g. CrazyStone? Is there a
> particular point where it adjusts who it thinks is winning?)
>
> Darren
>
> _______________________________________________
> Computer-go mailing list

> Compu...@computer-go.org <mailto:Compu...@computer-go.org>
> http://computer-go.org/mailman/listinfo/computer-go

[Michael Markefka]

I find it interesting that right until he ends his review, Antti only
praises White's moves, which are the human ones. When he stops, he
even considers a win by White as basically inevitable.

Now Fan Hui either blundered badly afterwards, or more promising, it
could be hard for humans to evaluate AlphaGo's play at this point
because they undervalue some it its choices. Which of course would be
similar to how some moves by the first world-beating chess AIs have
been treated by human experts.

AlphaGo might be even more of a wild card than it seems.

Also, on another note, that Google set up those Sedol games makes me
assume that they are convinced of actually succeeding. The Fan Hui
matches have been months ago, and AlphaGo will have spent that time
learning, and when the matches come around they will probably throw A
LOT of processing power at Sedol. I don't think they would try for so
much public reach to then fail and be associated with failure and
hybris.

[Xavier Combelle]

2016-01-28 12:23 GMT+01:00 Michael Markefka <michael....@gmail.com>:

I find it interesting that right until he ends his review, Antti only
praises White's moves, which are the human ones. When he stops, he
even considers a win by White as basically inevitable.

White moves are the AI ones, check the players

[Michael Markefka]

That would make my writing nonsense of course. :)

Thanks for the pointer.

[Darren Cook]

> here a comment by Antti Törmänen
> http://gooften.net/2016/01/28/the-future-is-here-a-professional-level-go-ai/

Thanks, exactly what I was looking for. He points out black 85 and 95
might be mistakes, but didn't point out any dubious white (computer)
moves. He picks out a couple of white moves as particularly good, e.g.
108, which is also an empty triangle: obviously AlphaGo isn't being held
back by any "good shape" heuristics ;-)

I hope he comments the other four games!

Darren

--
Darren Cook, Software Researcher/Developer
My new book: Data Push Apps with HTML5 SSE
Published by O'Reilly: (ask me for a discount code!)
http://shop.oreilly.com/product/0636920030928.do
Also on Amazon and at all good booksellers!

[Brian Sheppard]

I would just mention that Maven/Scrabble truncated rollouts are not comparable to Go/MCTS truncated rollouts. An evaluation function in Scrabble is readily at hand, because scoring points is hugely correlated with winning. There is no evaluation function for Go that is readily at hand.

There have been some efforts at whole-board evaluation in Go. Maybe NeuroGo was the earliest really cool demonstration. But I never saw anything that gave me confidence that the approach could work when embedded in an MCTS framework. I am blown away.

["Ingo Althöfer"]

Hello Tian,

the following is confidential, only for you.

I do not know, how the situation with your go project is.
But please let me know if I can help you in some way or another
to convince your boss that further work on computer-go makes
sense for facebook.

Best regards, Ingo.

[Josef Moudrik]

Strictly confidential, I dare say!

Dne st 2. 3. 2016 21:04 uživatel "Ingo Althöfer" <3-Hirn...@gmx.de> napsal: