The training of convolutional neural networks (CNN) is "Challenge yourself to beat Yesterday", and 30xxx network is doing her own job amazingly, currently 7200+ self elo and ongoing..
https://ibb.co/fpiVfV
P.S, Testers say 30xx network is still -100 to -150 elo beyond 11248.
Lco road to 8000 or 10000! self play elo.
Moderators: hgm, Rebel, chrisw
-
- Posts: 708
- Joined: Mon Jan 16, 2012 6:34 am
-
- Posts: 565
- Joined: Thu Nov 13, 2014 12:03 pm
Re: Lco road to 8000 or 10000! self play elo.
Leela has a promising career in politics if chess fails
-
- Posts: 50
- Joined: Mon Sep 10, 2018 10:15 am
- Location: Denmark
- Full name: Alexander Spence
Re: Lco road to 8000 or 10000! self play elo.
Mah!
I played a mini-match Lc0 Network 7200 elo Vs Lco Network N. 11149 elo 4851...
Four games, 30 sec to move
Lc0 with old NN is stronger
I played a mini-match Lc0 Network 7200 elo Vs Lco Network N. 11149 elo 4851...
Four games, 30 sec to move
Lc0 with old NN is stronger
-
- Posts: 2559
- Joined: Fri Nov 26, 2010 2:00 pm
- Location: Czech Republic
- Full name: Martin Sedlak
Re: Lco road to 8000 or 10000! self play elo.
This is hilarious, play 1 game next time.
I wish I knew that it's sufficient to test with just a handful of games, that would've saved me tremendous amount of time.
Martin Sedlak
-
- Posts: 4190
- Joined: Wed Nov 25, 2009 1:47 am
Re: Lco road to 8000 or 10000! self play elo.
The only hilarious thing is Lc0 self-testing.
-
- Posts: 10948
- Joined: Wed Jul 26, 2006 10:21 pm
- Full name: Kai Laskos
Re: Lco road to 8000 or 10000! self play elo.
To check the truthfulness of 2300+ Elo difference, even 3 games can be more than enough for high confidence of its falsification.
-
- Posts: 50
- Joined: Mon Sep 10, 2018 10:15 am
- Location: Denmark
- Full name: Alexander Spence
Re: Lco road to 8000 or 10000! self play elo.
Obviously it is a simple test, I do not pretend it is super precise, it is only indicative
I do not have time to play lots of games
But Lc0 plays against himself doing similar results (+3 = 1 -0)
75% of the games win, if they were tied, I would agree to say that we need to play a lot more games
But with these results I think the conclusions are clear
However I did do several mini-matches of 4 games with the latest NN very high and different from each other, but the old NN wins every time
-
- Posts: 4889
- Joined: Thu Mar 09, 2006 6:34 am
- Location: Pen Argyl, Pennsylvania
Re: Lco road to 8000 or 10000! self play elo.
Fake news....Nay Lin Tun wrote: ↑Fri Nov 23, 2018 8:47 am
P.S, Testers say 30xx network is still -100 to -150 elo beyond 11248.
anyone can verify or disprove your claims after even just a few games, I believe it's actually worse...
Code: Select all
Rank Name Rating Δ + - # Σ Σ% W L D W% =% OppR
---------------------------------------------------------------------------------------------------------
1 Lc0 v0.19.0 11261 3198 0.0 84 84 50 38.5 77.0 33 6 11 66.0 22.0 3002
2 Lc0 v0.19.0 31493 3002 195.7 84 84 50 11.5 23.0 6 33 11 12.0 22.0 3198
---------------------------------------------------------------------------------------------------------
-
- Posts: 4319
- Joined: Tue Apr 03, 2012 4:28 pm
Re: Lco road to 8000 or 10000! self play elo.
I hope this isn't happening ....
run program ....
leela self_play elo has liftoff, Houston! Elo rating=6250
Code: Select all
import random
program_new_trained_version_elo = 0
program_old_version_elo = 0
current_best_leela_elo = 0
we_didnt_get_bored_yet = 0
while (we_didnt_get_bored_yet < 2500):
# play a match, the programs are the same because learning stalled, but the results are always a bit random ....
random_result = (random.randint(0,1))
if (random_result == 1):
# if new trained version happens to be randomly better, like 50% of the time ...
# keep new trained version as leela, and increase the elo a bit (because Leela won this one)
program_new_trained_version_elo += 5
current_best_leela_elo = program_new_trained_version_elo
# else:
# new trained version happens to not be randomly better
# don't update leela
we_didnt_get_bored_yet += 1
# we got bored ....
print('leela self_play elo has liftoff, Houston! Elo rating=' + str(current_best_leela_elo))
leela self_play elo has liftoff, Houston! Elo rating=6250
-
- Posts: 50
- Joined: Mon Sep 10, 2018 10:15 am
- Location: Denmark
- Full name: Alexander Spence
Re: Lco road to 8000 or 10000! self play elo.
YesMikeB wrote: ↑Fri Nov 23, 2018 8:38 pm Fake news....
anyone can verify or disprove your claims after even just a few games, I believe it's actually worse...
custom openings that may exaggerate Elo differences ( due to the unbalance nature of the openings )..Code: Select all
Rank Name Rating Δ + - # Σ Σ% W L D W% =% OppR --------------------------------------------------------------------------------------------------------- 1 Lc0 v0.19.0 11261 3198 0.0 84 84 50 38.5 77.0 33 6 11 66.0 22.0 3002 2 Lc0 v0.19.0 31493 3002 195.7 84 84 50 11.5 23.0 6 33 11 12.0 22.0 3198 ---------------------------------------------------------------------------------------------------------
I used NN 11149, I think it is slightly higher than 11261 and 11248