Yes, node count is important, not the GPU. Anything in the range of 1k - 1000k nodes per move benefits from Kiudee settings, and with most of the nets (well, not too weird ones).kasinp wrote: ↑Mon Jan 27, 2020 10:50 pmAre these setting unique to RTX cards? Would I benefit from trying them on a GTX card?Kiudee wrote: ↑Wed Jan 22, 2020 6:02 pmThe tuned settings were done using a mix of short time controls using the T58 (58613) network. These settings should be approximately optimal for all time controls with <20k nodes / move. Even though the CPuct scaling curve is not uniquely determined on slower time controls, it can perform reasonably well on time controls up to <100k nodes, but is likely slightly suboptimal there (though it seemed to perform quite well for T60 on CCC's time control). For 1M+ nodes: no guarantees
I am currently in the process of writing a distributed tuning framework for lc0, which we then could utilize to tune slower time controls than before. These parameters will then likely be better candidates as new defaults, than the currently circulating ones.
Thank you
Peter
Lc0: Kiudee setting is very strong
Moderators: hgm, Rebel, chrisw
-
- Posts: 10948
- Joined: Wed Jul 26, 2006 10:21 pm
- Full name: Kai Laskos
Re: Lc0: Kiudee setting is very strong
-
- Posts: 251
- Joined: Sat Dec 02, 2006 10:47 pm
- Location: Toronto
- Full name: Peter Kasinski
Re: Lc0: Kiudee setting is very strong
Thank you, your explanation helps. I will certainly give it a try.Laskos wrote: ↑Tue Jan 28, 2020 12:38 amYes, node count is important, not the GPU. Anything in the range of 1k - 1000k nodes per move benefits from Kiudee settings, and with most of the nets (well, not too weird ones).kasinp wrote: ↑Mon Jan 27, 2020 10:50 pmAre these setting unique to RTX cards? Would I benefit from trying them on a GTX card?Kiudee wrote: ↑Wed Jan 22, 2020 6:02 pmThe tuned settings were done using a mix of short time controls using the T58 (58613) network. These settings should be approximately optimal for all time controls with <20k nodes / move. Even though the CPuct scaling curve is not uniquely determined on slower time controls, it can perform reasonably well on time controls up to <100k nodes, but is likely slightly suboptimal there (though it seemed to perform quite well for T60 on CCC's time control). For 1M+ nodes: no guarantees
I am currently in the process of writing a distributed tuning framework for lc0, which we then could utilize to tune slower time controls than before. These parameters will then likely be better candidates as new defaults, than the currently circulating ones.
Thank you
Peter
Peter
-
- Posts: 74
- Joined: Wed Dec 04, 2019 11:25 am
- Full name: Prasanna Bandihole
Re: Lc0: Kiudee setting is very strong
There was an earlier suggestion to include changing "Shortsightedness" value. It is suggested in the Github to use a value of 0.02. Someone had tested with 0.05 with good result.
-
- Posts: 29
- Joined: Tue Feb 02, 2010 10:12 pm
- Location: Germany
- Full name: Karlson Pfannschmidt
Re: Lc0: Kiudee setting is very strong
I did a tune of only ShortSightedness. The parameter interacts with the other parameters like CPuct, etc, which is why I will do a proper joint tune later. The result of the isolated tune is here:
A value of 0.05 is significantly worse than 0.02 or the other optima and cannot be recommended.
-
- Posts: 29
- Joined: Tue Feb 02, 2010 10:12 pm
- Location: Germany
- Full name: Karlson Pfannschmidt
Re: Lc0: Kiudee setting is very strong
The settings are not unique to RTX cards - in fact it was tuned on a GTX 1080.kasinp wrote: ↑Mon Jan 27, 2020 10:50 pmAre these setting unique to RTX cards? Would I benefit from trying them on a GTX card?Kiudee wrote: ↑Wed Jan 22, 2020 6:02 pmThe tuned settings were done using a mix of short time controls using the T58 (58613) network. These settings should be approximately optimal for all time controls with <20k nodes / move. Even though the CPuct scaling curve is not uniquely determined on slower time controls, it can perform reasonably well on time controls up to <100k nodes, but is likely slightly suboptimal there (though it seemed to perform quite well for T60 on CCC's time control). For 1M+ nodes: no guarantees
I am currently in the process of writing a distributed tuning framework for lc0, which we then could utilize to tune slower time controls than before. These parameters will then likely be better candidates as new defaults, than the currently circulating ones.
Thank you
Peter
-
- Posts: 3657
- Joined: Wed Nov 18, 2015 11:41 am
- Location: hungary
Re: Lc0: Kiudee setting is very strong
I wait curiously the result of your works but I am afraid there are no universally "best" parameters because the effectiveness of parameters depend not only from Leela and the used net and hardware but the behavior of the opponent also.Kiudee wrote: ↑Wed Jan 22, 2020 6:02 pm ...
I am currently in the process of writing a distributed tuning framework for lc0, which we then could utilize to tune slower time controls than before. These parameters will then likely be better candidates as new defaults, than the currently circulating ones.
You can determine "best" parameters for CCC final where the GPUs, the used net and the opponent (Stockfish, mainly) are given but we, common Leela and net user have different GPUs, different nets and we use Leela against different opponents. So I think we should be satisfied if you can find in some measure "better" but universal parameters instead of the "best" parameters.
Basing on my experience the most of the default parameters of Leela are rather good for general using.
-
- Posts: 29
- Joined: Tue Feb 02, 2010 10:12 pm
- Location: Germany
- Full name: Karlson Pfannschmidt
Re: Lc0: Kiudee setting is very strong
I agree that for many parameters you can find specific values optimal for a very specific time control. That being said, for default parameters it still makes sense to provide a configuration, which works well for a wide range of time controls (we can’t expect a typical user to be an engine tuning expert, who (1) knows with which parameters to fiddle and (2) knows that thousands of games are necessary to find a better configuration).corres wrote: ↑Tue Jan 28, 2020 10:35 amI wait curiously the result of your works but I am afraid there are no universally "best" parameters because the effectiveness of parameters depend not only from Leela and the used net and hardware but the behavior of the opponent also.Kiudee wrote: ↑Wed Jan 22, 2020 6:02 pm ...
I am currently in the process of writing a distributed tuning framework for lc0, which we then could utilize to tune slower time controls than before. These parameters will then likely be better candidates as new defaults, than the currently circulating ones.
You can determine "best" parameters for CCC final where the GPUs, the used net and the opponent (Stockfish, mainly) are given but we, common Leela and net user have different GPUs, different nets and we use Leela against different opponents. So I think we should be satisfied if you can find in some measure "better" but universal parameters instead of the "best" parameters.
Basing on my experience the most of the default parameters of Leela are rather good for general using.
In general, we try to design parameters which naturally adapt to the given time control. The cpuct scaling curve (as defined by cpuct, cpuctbase and cpuctfactor) tries to accomplish exactly that: increasing cpuct when more and more visits are coming in. And many people forget that a lower cpuct is not only an improvement in the root node, but does improve the quality of the evaluations in the complete tree (thus it even improves the search for LTC). Still, I think that the scaling curve is not yet optimal for time controls with 100k+ nodes per move.
-
- Posts: 325
- Joined: Sat Nov 30, 2013 9:51 am
- Location: Canada
- Full name: Jason Coombs
Re: Lc0: Kiudee setting is very strong
Off topic:
Is "Fpu" the same as "FpuValue" ? because I don't see "Fpu" in the uci options. Thanks
Is "Fpu" the same as "FpuValue" ? because I don't see "Fpu" in the uci options. Thanks
-
- Posts: 10948
- Joined: Wed Jul 26, 2006 10:21 pm
- Full name: Kai Laskos
-
- Posts: 256
- Joined: Wed Oct 02, 2013 12:36 am
Re: Lc0: Kiudee setting is very strong
I had a run in my LTC chess engine ladder, (90 min 45 sec/per move) using latest Lc0 version as of jan 28, with blass backserver (no gpu).
Lco works very well up to middle game, approx up to move 35, then it loses eggregiously afterwards. I suspect configuration using blass must be quite different. Any suggestions for blass-based configuration?
Lco works very well up to middle game, approx up to move 35, then it loses eggregiously afterwards. I suspect configuration using blass must be quite different. Any suggestions for blass-based configuration?