[Computer-go] 1st Go Denou-sen

Hideki Kato hideki_katoh at ybb.ne.jp
Tue Feb 11 18:08:00 PST 2014


Attached.

Some comments for the games.

1st game: W move 18 should be E3 and the game was over (by Meien O 
9p).  W move 32 (D5) was bad (ditto) but Hirata 3p claimed he worried 
B C5.  #Professionals spent much time for this game and concluded Zen 
had a chance to win (I couldn't follow ;- .  Perhaps, move 35 (E8) 
should be C8?  Comments welcome).

2nd game: Almost perfect win by B.  Chang 8p had studied this opening a 
lot and had a big confidence.  Ohashi 6p (well-known 9x9 expert) was 
very glad because the opening until move 7 practically get concluded as 
B+0.5, thanks to Chang.  W E5 (6th) was bad.  #This could be included 
in opening books.  Chang also commented he couldn't noticed W F7 
(8th) by himself, which is a very good move.  So, without pre-study, 
he would be surprised and was hard to win.

3rd game: W F7 was good (perhaps best).  W H2 (16th) surprised Hirata 
3p (he commented after the game) but was suspicious.  Perhaps, F4 was 
better but either looks lead B win.  Ie, W already behind here.  W E5 
was bad?   I'm not sure.  Anyway, B E4 (17th) was good which Zen found 
after played H2.  At H2, Zen forecast F4 and (I guess) there was 1 pt 
(or less) difference by W E4.  W had a chance to play G5 and start a ko 
at F4.  After B G5 (21st), W had no chance.  Almost perfect win by B.

4th game: Some people think Chang intentionally lead a complicated 
game (not confirmed :) but Zen did a big mistake (B C4) and lost.  
Weak programs commonly played D8 (correct) but Zen found he could kill 
bottom W (wrong).  B D8 lead even positions but C4 lead a sure win if 
bottom W could be killed.  So, unfortunatelly, this was the natural 
choice for Zen.  Two move later (B C3), Zen noticed W could live and 
winrate dropped to 30%.  Alas, too late!  
#W E2 (14th) is a well known tesuji to live on sides but Zen has no 
special procedure for this shape in the simulation.  If the cpu was 10 
time faster, Zen could play D8 :).

Overall: I found that when Zen was behind, professionals played the 
same moves as Zen's forecast.  This could mean professionals played 
without mistakes.  MCTS bots do some mistakes due to the stochastic 
behavior, which is a essential (i.e. cannot avoid) weak point of 
the algorithm.  More effort would be necessary to build better opening 
books, anyway.

Hideki

Rémi Coulom: <FEC692C0-7750-46BA-AD2D-7718ED4485FA at free.fr>:
>Hi Hideki,
>
>Thanks for your comments. Are the game records available?
>
>Rémi
>
>On 11 févr. 2014, at 22:19, Hideki Kato <hideki_katoh at ybb.ne.jp> wrote:
>
>> Professinals can play with no or few mistakes on 9x9 but MC bots play 
>> with more mistakes (perhaps one per 10 to 20 moves?) due to the 
>> randomness.  At the two-years-ago event, Zen vs 3 pros, both did many 
>> mistakes and Zen had several chances to win.  At this event, however, 
>> Cho 8p played almost perfectly because he studied 9x9 Go so long (more 
>> than 10 hours) and Zen had no chances.  Hirata 3p (younger pro) did few 
>> mistakes at the 1st game but Zen couldn't take the chance because it was 
>> a very narrow sequence of moves.  As a conclusion, no-miss-play is 
>> required to beat serious professionals on 9x9 :).
>> 
>> Hideki
>> 
>> Hiroshi Yamashita: <6582F7E197A74985AC2DEB659A9F7753 at i3540>:
>>>> It seems Zen lost the 4 games:
>>> 
>>> Yes, Zen lost 4 games in 9x9.
>>> 
>>> Chang Li Yu,   8d Pro - Zen  2-0
>>> Hirata Tomoya, 3d Pro - Zen  2-0
>>> 
>>> Here is a movie news.
>>> 
>>> Go Denou-sen, pro win decisively (in Japanese)
>>> http://www3.nhk.or.jp/news/html/20140211/k10015174611000.html
>>> 
>>> Hiroshi Yamashita
>>> 
>>> _______________________________________________
>>> Computer-go mailing list
>>> Computer-go at dvandva.org
>>> http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
>> -- 
>> Hideki Kato <mailto:hideki_katoh at ybb.ne.jp>
>> _______________________________________________
>> Computer-go mailing list
>> Computer-go at dvandva.org
>> http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
>
>_______________________________________________
>Computer-go mailing list
>Computer-go at dvandva.org
>http://dvandva.org/cgi-bin/mailman/listinfo/computer-go
-- 
Hideki Kato <mailto:hideki_katoh at ybb.ne.jp>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: 20140211.zip
Type: application/octet-stream
Size: 2090 bytes
Desc: not available
URL: <http://computer-go.org/pipermail/computer-go/attachments/20140212/2fcc17f3/attachment.obj>


More information about the Computer-go mailing list