| Life In 19x19 http://lifein19x19.com/ |
|
| so basically leela zero has made 0 progress in the last 4mo http://lifein19x19.com/viewtopic.php?f=18&t=16227 |
Page 2 of 2 |
| Author: | bernds [ Sat Nov 24, 2018 9:31 am ] |
| Post subject: | Re: so basically leela zero has made 0 progress in the last |
Relevant discussion elsewhere: https://www.reddit.com/r/cbaduk/comments/9zv4v1/leelazero_40b_networks_are_overfit/ |
|
| Author: | ez4u [ Sat Nov 24, 2018 10:58 pm ] |
| Post subject: | Re: so basically leela zero has made 0 progress in the last |
bernds wrote: Relevant discussion elsewhere: https://www.reddit.com/r/cbaduk/comments/9zv4v1/leelazero_40b_networks_are_overfit/ This is a very interesting report! The follow up posting on the LZ site is here: https://github.com/gcp/leela-zero/issues/2044 |
|
| Author: | sorin [ Sun Nov 25, 2018 1:53 pm ] |
| Post subject: | Re: so basically leela zero has made 0 progress in the last |
ez4u wrote: The follow up posting on the LZ site is here: https://github.com/gcp/leela-zero/issues/2044 To summarize the github thread: the claim that starting with pre-set openings somehow makes 40 block LZ weaker than 15 block is wrong. The error was due to using incorrect setup params, which somehow seem to affect pre-set games but not regular games. Phew, the world is sane again 
|
|
| Author: | Vargo [ Fri Nov 30, 2018 6:25 am ] |
| Post subject: | Re: so basically leela zero has made 0 progress in the last |
The networks do get better, even at short time parity (~1 sec/move) , see HERE |
|
| Author: | roy7 [ Mon Dec 24, 2018 8:23 am ] |
| Post subject: | Re: so basically leela zero has made 0 progress in the last |
bernds wrote: I was curious, so I was trying to run something similar yesterday. While I got a similar result, the games all looked identical up to move 40 or so. I think I saw a comment in the github by someone else who ran comparisons that one of the tools people sometimes use to run these head to head matches (twogtp?) doesn't restart the engine between games, it just uses the existing engine in memory and sets up the new board/etc and begins playing the new game. The issue with this at the time (perhaps a recent pull request addressed it) is that the NN eval cache doesn't flush just because you cleared the board. Thus you'll have exactly the same results of every early NN eval each game you play. (Normally symmetry causes some small random noise, but once cached the value never changes.) |
|
| Author: | hydrogenpi7 [ Sat Jan 19, 2019 12:04 pm ] |
| Post subject: | Re: so basically leela zero has made 0 progress in the last |
So basically I was wrong, make that half a year zero net effective progress ... |
|
| Page 2 of 2 | All times are UTC - 8 hours [ DST ] |
| Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group http://www.phpbb.com/ |
|