It is currently Tue Mar 28, 2017 10:21 pm

All times are UTC - 8 hours [ DST ]




Post new topic Reply to topic  [ 105 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6
Author Message
Offline
 Post subject: Re: Leela has a new version
Post #101 Posted: Mon Feb 27, 2017 11:14 am 
Lives in gote

Posts: 365
Liked others: 25
Was liked: 103
GD Posts: 10
Looking forward to leela-0.9.2-4c performance in cgos.
Maybe we have to wait until v.1.0 for public release.

Top
 Profile  
 
Offline
 Post subject: Re: Leela has a new version
Post #102 Posted: Sat Mar 04, 2017 2:39 pm 
Beginner

Posts: 8
Liked others: 0
Was liked: 2
Rank: 5k


There was an interesting game today on KGS between Ilya Shikshin and Leela.
In fact Leela seemed to be ahead and in control during the whole game but it miserably failed a few moves before the end and resigned there.

The position that interests us really begins at move 188 but the climax is move 191. D17

When analysing the game with my home version 0.9.0, I noticed that in every variation displayed by Leela before this move, then D17 was systematically assumed to be answered by E17, so that an eventual ko at Q19 could be handled.

As we can see at move 191. from this screenshot, Leela only considers 3 potential answers and still estimates his chances of success very high :

Image

But Ilya played next 192. M19 and Leela realized this was hopeless and winning % droped from 70% to less than 10% !

I ran the position in various engines, Zen 9.7 is also overseeing this move for 30s before it appears effectively in its candidates move list and almost immediately realizes that it is a dreadful threat. Surprisingly the 'allmighty' Fuego plays M19 blindly and instantly.

I was wondering how such powerful engines not only cannot see this immediately but do not even consider a move as important as an atari onto a big group ?! I let Leela run for more than 5 minutes, and it never considered candidates moves other than the 3 above.

For engines to oversee a threat because it is behinf the search horizon is a thing, but to not even assess a potential atari seems quite a big flaw in the design.

What do you think ? Are you able to reproduce this on other versions of Leela ?

Top
 Profile  
 
Offline
 Post subject: Re: Leela has a new version
Post #103 Posted: Sun Mar 05, 2017 7:23 am 
Dies in gote

Posts: 20
Liked others: 0
Was liked: 10
Rank: OGS 7 kyu
Hi,

I just followed your link from my other thread.
Quote:
Surprisingly the 'allmighty' Fuego plays M19 blindly and instantly.
In fact, even the venerable GnuGo would play M19

Quote:
Are you able to reproduce this on other versions of Leela ?
Same issue with Leela 6.2

Top
 Profile  
 
Offline
 Post subject: Re: Leela has a new version
Post #104 Posted: Sun Mar 05, 2017 11:51 am 
Dies in gote

Posts: 28
Liked others: 1
Was liked: 4
Rank: KGS 2 D
zwim wrote:

But Ilya played next 192. M19 and Leela realized this was hopeless and winning % droped from 70% to less than 10% !



After viewed the games with Haylee, I observed similar behaviors of Leela. It has a strong tendency to follow wherever the opponent plays locally. Unlike zen, it doesn't play save (defensively) even when it is far ahead. So after battles around the board from here to there, it will forget what the most important move is globally.

Top
 Profile  
 
Offline
 Post subject: Re: Leela has a new version
Post #105 Posted: Sun Mar 05, 2017 2:42 pm 
Dies with sente

Posts: 77
Liked others: 0
Was liked: 23
Rank: NR
zwim wrote:
I was wondering how such powerful engines not only cannot see this immediately but do not even consider a move as important as an atari onto a big group ?! I let Leela run for more than 5 minutes, and it never considered candidates moves other than the 3 above.

For engines to oversee a threat because it is behinf the search horizon is a thing, but to not even assess a potential atari seems quite a big flaw in the design.

the reason dcnn engines sometimes appear to not consider good moves that are obvious to people is that whereas dcnn's so-called intuition (sic) "sees" convolutions of pattern fragments, they don't "see" commonsense Go concepts such as groups, eyes, etc. it was this fundamental flaw that led Alphago to miss the "obvious" defence to Lee's magic wedge which didn't work.

that's why i advocate combining commonsense and dcnn in a jury (or committee discussion rounds as suggested by UpWarrior)
http://lifein19x19.com/viewtopic.php?f=18&t=14049

Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 105 posts ]  Go to page Previous  1, 2, 3, 4, 5, 6

All times are UTC - 8 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 3 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group