Posts : 2603 Join date : 2020-11-17 Location : Netherlands
Subject: A new Benjamin ? Tue Sep 17, 2024 1:08 pm
Benjamin 1.0 was the gambit version of ProDeo.
I am now working on a possible Benjamin 2.0 version meant to be the gambit version of Rebel.
Elo is not the end to the means, playing style is, and Stefan's EAS tool is still the best around to measure aggressiveness. This will be a project for months generating data with occasional updates.
Code:
Net size 850M | Net size 1.7B | Epoch EAS | Epoch EAS | 8 267044 | 16 334680 | 16 234491 | 32 282878 | | 39 280698 |
Currently the 1.7B version shows a loss of 140 elo in comparison to Rebel 16.3, I have hope the loss will be limited to ~100 elo when I have generated 10B of data, estimated hopefully still in 2024
Surprisingly I noticed the new generated data (so far) doesn't bite the large 25.5B Rebel 16.3 data, it even gave a small improvement and also a higher EAS score. But there are a lot of maybe's here.
All in all I have the fun back in NNUE programming after the discord debacle. I grew up in the 80's with originality, originality and originality and I prefer to stick with that, but maybe I am old fashioned.
Last edited by Admin on Sat Oct 05, 2024 8:46 am; edited 1 time in total
Tibono, Eelco, pohl4711, Ipmanchess and Ghppn like this post
pohl4711
Posts : 150 Join date : 2022-03-01 Location : Berlin
Subject: Re: A new Benjamin ? Wed Sep 18, 2024 7:06 am
Admin wrote:
Benjamin 1.0 was the gambit version of ProDeo.
I am now working on a possible Benjamin 2.0 version meant to be the gambit version of Rebel.
Elo is not the end to the means, playing style is, and Stefan's EAS tool is still the best around to measure aggressiveness. This will be a project for months generating data with occasional updates.
Code:
Net size 850M | Net size 1.7B | Epoch EAS | Epoch EAS | 8 267044 | 16 334680 | 16 234491 | 32 282878 | | 39 280698 |
EAS scores of 280k-334k are very impressive! Of course, such high EAS scores must cost Celos. Playing high risky style is nice - but you have to pay a price... Celos have to be invested - no way to avoid this. Only question is, how many Celos. IMHO a Celo-investment below -150 is already a good trade.
I keep my fingers crossed! Cant wait to see a new Rebel EAS!!!
Admin, Mclane and Ghppn like this post
Admin Admin
Posts : 2603 Join date : 2020-11-17 Location : Netherlands
Subject: Re: A new Benjamin ? Sat Sep 21, 2024 8:45 pm
Fun...
Code:
Net size 850M | Net size 1.7B | Epoch EAS | Epoch EAS | 8 267044 | 16 334680 | 16 234491 | 32 282878 | | 39 280698 |
I have been playing a 3000+ game match between Patricia and the epoch 39 version, the EAS :
Gone is Patricia's 400K score Gone is Benjamin's 300K score
Oh, then the match score :
Code:
No. Name Win Draw Loss Unf. Score Games % ---------------------------------------------------------- 1 Benjamin-E39 +1823 =1115 -146 *0 2380.5 3084 77.2% 2 patricia-3.01 +146 =1115 -1823 *0 703.5 3084 22.8%
Ouch, I thought Patricia was a lot stronger.
This result has nothing to do with the lower strength of Patricia. This is (for me) a well known effect: When 2 very aggressive playing engines just play each other, the EAS-scores are much smaller for both engines, compared to the EAS-scores of these engines, when playing vs a pool of "normal" playing engines. And this is not surprising, IMHO, because both steal EAS from the opponent: A very aggressive engines plays a lot of speculative sacs, which often go wrong and lead to a loss. But in these games, the (aggressive) opponent has no chance, to play an own sac. So, the EAS-scores are much smaller for both engines as a result (lost games with a sac from the loosing color, are not counted by the EAS-tool) You should play 2 gauntlets (of Patricia and Benjamin) versus a pool of normal engines and compare the EAS-results of these 2 gauntlet-testruns...
Ghppn likes this post
Admin Admin
Posts : 2603 Join date : 2020-11-17 Location : Netherlands
Subject: Re: A new Benjamin ? Mon Sep 23, 2024 9:51 am
Yeah, they bite each other, where two fight for EAS both got damaged, fun to watch, no need to repeat the experiment.
Admin Admin
Posts : 2603 Join date : 2020-11-17 Location : Netherlands
Subject: Re: A new Benjamin ? Mon Sep 23, 2024 11:12 am
Started a new training session for a possible REBEL EAS 2.0
Trainer states - Training positions: 27272727272
But I am not superstitious
Damir Desevac, Dio and Ghppn like this post
Admin Admin
Posts : 2603 Join date : 2020-11-17 Location : Netherlands
Subject: Re: A new Benjamin ? Thu Sep 26, 2024 7:07 pm
matejst, Tibono, Damir Desevac, Eelco and Ghppn like this post
texium
Posts : 124 Join date : 2022-07-19
Subject: Re: A new Benjamin ? Fri Sep 27, 2024 8:36 pm
i sincerely think adding human data helps, and makes engines stronger. for example, when i added some net trained on only human games to stockfish classical, it makes it stronger, maybe filling gaps. human games online are very sharp and provide much training data to help.
for example, here's a leela net at only 10k steps trained on human games, that plays more aggressive than any aggressive engine at its elo, while being ridiculously sharp for leela. [pgn][Event "new nets test"] [Site "Lucas Chess R 2.14l"] [Date "2024.09.25"] [Round "?"] [White "Tao 5.6"] [Black "Leela Tal Blitz 10k steps"] [Result "0-1"] [WhiteElo "2491"] [BlackElo "2300"] [ECO "A00"] [Opening "Van Geet Opening: Napoleon Attack"] [TimeControl "12+0"] [Termination "Mate"] [PlyCount "120"] [Link "https://www.chess.com/analysis/game/pgn/26kfbJDzpW?tab=analysis&move=47"]
[/pgn] what it learned was to play for mate, organizing its pieces and going for the win. it feels like if tal played online blitz today. i've had it run on lichess, and it equalized a completely lost position, but lost in the endgame.
Subject: Re: A new Benjamin ? Thu Oct 03, 2024 3:35 pm
texium wrote:
i sincerely think adding human data helps, and makes engines stronger. for example, when i added some net trained on only human games to stockfish classical, it makes it stronger, maybe filling gaps. human games online are very sharp and provide much training data to help.
for example, here's a leela net at only 10k steps trained on human games, that plays more aggressive than any aggressive engine at its elo, while being ridiculously sharp for leela. [pgn][Event "new nets test"] [Site "Lucas Chess R 2.14l"] [Date "2024.09.25"] [Round "?"] [White "Tao 5.6"] [Black "Leela Tal Blitz 10k steps"] [Result "0-1"] [WhiteElo "2491"] [BlackElo "2300"] [ECO "A00"] [Opening "Van Geet Opening: Napoleon Attack"] [TimeControl "12+0"] [Termination "Mate"] [PlyCount "120"] [Link "https://www.chess.com/analysis/game/pgn/26kfbJDzpW?tab=analysis&move=47"]
[/pgn] what it learned was to play for mate, organizing its pieces and going for the win. it feels like if tal played online blitz today. i've had it run on lichess, and it equalized a completely lost position, but lost in the endgame.
Could you point me to where I can find this interesting Leela Tal 10k steps net?
Ghppn likes this post
texium
Posts : 124 Join date : 2022-07-19
Subject: Re: A new Benjamin ? Fri Oct 04, 2024 11:01 am
that was a bad pool, filled with aggressive engines, which i know lowers eas rating. here i tested using engines from ccrl, assuming 26-2700, and ccrl testing standards. i did 20 games per engine. 10 on each color, ccrl doesnt even do that many. it only had 18 draws out of 120 games. though with 46 wins, i doubt it would get much lower with a larger sample
Ghppn likes this post
texium
Posts : 124 Join date : 2022-07-19
Subject: Re: A new Benjamin ? Fri Oct 04, 2024 11:02 am
Brendan wrote:
texium wrote:
And this game, is just beautiful.
Could you point me to where I can find this interesting Leela Tal 10k steps net?
i can send it to you privately.
Damir Desevac
Posts : 322 Join date : 2020-11-27 Age : 43 Location : Denmark
Subject: Re: A new Benjamin ? Fri Oct 04, 2024 4:44 pm
I too am interested. Can you send me a PM ?
Ghppn likes this post
Griffin
Posts : 1 Join date : 2024-10-01
Subject: Re: A new Benjamin ? Fri Oct 04, 2024 5:33 pm
texium wrote:
i can send it to you privately.
If possible. I also want to get these weights.
Ghppn likes this post
Brendan
Posts : 399 Join date : 2020-11-18 Age : 40
Subject: Re: A new Benjamin ? Sat Oct 05, 2024 2:31 am
texium wrote:
Brendan wrote:
texium wrote:
And this game, is just beautiful.
Could you point me to where I can find this interesting Leela Tal 10k steps net?
i can send it to you privately.
Please do. Would be much appreciated.
Ghppn likes this post
texium
Posts : 124 Join date : 2022-07-19
Subject: Re: A new Benjamin ? Yesterday at 9:12 am
Subject: Re: A new Benjamin ? Yesterday at 9:51 am
150 wins and idk how many draws
Tibono
Posts : 29 Join date : 2020-11-27 Location : France
Subject: Re: A new Benjamin ? Yesterday at 1:35 pm
A logo for Leela TAL (thks, Texium)
Admin Admin
Posts : 2603 Join date : 2020-11-17 Location : Netherlands
Subject: Re: A new Benjamin ? Yesterday at 4:07 pm
I see Leela-10 is a NN, can it be used in the CPU version of Leela?
Dio
Posts : 217 Join date : 2021-08-28
Subject: Re: A new Benjamin ? Yesterday at 4:53 pm
yes, this is possible. However, a NN that was only trained with 10,000 training games is very weak. I doubt the origin of this NN.
Last edited by Dio on Thu Oct 17, 2024 5:03 pm; edited 1 time in total
texium
Posts : 124 Join date : 2022-07-19
Subject: Re: A new Benjamin ? Yesterday at 5:02 pm
Dio wrote:
yes, this is possible. However, a NN that was only trained with 16,000 training games is very weak. I doubt the origin of this NN.
Actually it's plenty strong considering it was only trained on human games to 10k steps. Badgyal has a hybrid of human and stockfish and was only 3400.
Admin Admin
Posts : 2603 Join date : 2020-11-17 Location : Netherlands
Subject: Re: A new Benjamin ? Yesterday at 5:03 pm
Dio wrote:
yes, this is possible. However, a NN that was only trained with 16,000 training games is very weak. I doubt the origin of this NN.
Well, that's one those keys for a high EAS
But at least with the CPU version one can play with a high concurrency.
Home 
