Lc0 soon to support chess960 ?

Discussion of anything and everything relating to chess playing software and machines.

Moderators: Harvey Williamson, bob, hgm

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Post Reply
Modern Times
Posts: 2604
Joined: Thu Jun 07, 2012 9:02 pm

Lc0 soon to support chess960 ?

Post by Modern Times » Sat Apr 18, 2020 8:44 am

v0.25.0-rc1
Training now outputs v5 format data. This passes the moves left information back to training. This also includes support for multiple sub formats,
including the existing standard, a new variant which can encode FRC960 castling, and also a further extension of that which tries to make training
data cannonical, so there aren't multiple positions that are trivially equivalent with different network inputs.
And the engine now has a UCI parameter UCI_chess960
.

Opinions expressed here are my own, and not necessarily those of the CCRL Group.

Kiudee
Posts: 27
Joined: Tue Feb 02, 2010 9:12 pm

Re: Lc0 soon to support chess960 ?

Post by Kiudee » Sat Apr 18, 2020 11:06 am

As a matter of fact, we already trained a Chess960 network to test the implementation.
Take a look at our list of networks and download for example the network 712574.
It supports both Chess960 and the new moves-left-head.

If you want to see the moves-left-head in action, try these additional settings:

Code: Select all

setoption name MovesLeftMaxEffect value 0.9
setoption name MovesLeftThreshold value 0.3
setoption name MovesLeftSlope value 0.009
setoption name MovesLeftQuadraticFactor value 1.0
setoption name MovesLeftConstantFactor value 0.0
With these settings, it will favor the quickest win whenever possible and you will most likely see much less shuffling in won positions.
Note, that this is a very small test run very early in its development. Try it against weaker opponents like Pedone etc.

edit: As a bonus, the GUI Nibbler now also supports Chess960

Modern Times
Posts: 2604
Joined: Thu Jun 07, 2012 9:02 pm

Re: Lc0 soon to support chess960 ?

Post by Modern Times » Sat Apr 18, 2020 4:43 pm

Thanks. Allie 0.5 did amazingly well for me just using the standard 42850 net, but I reckon there is +100 to 150 Elo to be gained on a chess960 trained net.
.

Opinions expressed here are my own, and not necessarily those of the CCRL Group.

Modern Times
Posts: 2604
Joined: Thu Jun 07, 2012 9:02 pm

Re: Lc0 soon to support chess960 ?

Post by Modern Times » Mon Apr 20, 2020 6:04 pm

Kiudee wrote:
Sat Apr 18, 2020 11:06 am
As a matter of fact, we already trained a Chess960 network to test the implementation.
Take a look at our list of networks and download for example the network 712574.
It supports both Chess960 and the new moves-left-head.
Yes I tried that net at chess960 and it wasn't great.

I'm currently running it against 591226 to establish a baseline, and we'll see how it improves from there.
.

Opinions expressed here are my own, and not necessarily those of the CCRL Group.

dkappe
Posts: 542
Joined: Tue Aug 21, 2018 5:52 pm
Full name: Dietrich Kappe

Re: Lc0 soon to support chess960 ?

Post by dkappe » Mon Apr 20, 2020 6:25 pm

Both MiniHuman and leela2200 on lichess play chess960 on lc0 0.24, so the capability has already been there.

Modern Times
Posts: 2604
Joined: Thu Jun 07, 2012 9:02 pm

Re: Lc0 soon to support chess960 ?

Post by Modern Times » Mon Apr 20, 2020 6:48 pm

dkappe wrote:
Mon Apr 20, 2020 6:25 pm
Both MiniHuman and leela2200 on lichess play chess960 on lc0 0.24, so the capability has already been there.
Amazing, I did not see that anywhere.
.

Opinions expressed here are my own, and not necessarily those of the CCRL Group.

Modern Times
Posts: 2604
Joined: Thu Jun 07, 2012 9:02 pm

Re: Lc0 soon to support chess960 ?

Post by Modern Times » Wed Apr 22, 2020 8:07 am

Kiudee wrote:
Sat Apr 18, 2020 11:06 am
As a matter of fact, we already trained a Chess960 network to test the implementation.
Take a look at our list of networks and download for example the network 712574.
It supports both Chess960 and the new moves-left-head.

<snip>
Note, that this is a very small test run very early in its development. Try it against weaker opponents like Pedone etc.

Yes I did try 712574 and it was pretty weak compared to what Lc0 is capable of in normal chess. But the first results are up with 591226

http://ccrl.chessdom.com/ccrl/404FRC/

More still to come.

If there is ever a trained 960 network it will be interesting to see the improvement. The effect would be earlier in the games; once they get to a certain point they are mostly indistinguishable from standard games .
.

Opinions expressed here are my own, and not necessarily those of the CCRL Group.

Hai
Posts: 548
Joined: Sun Aug 04, 2013 11:19 am

Re: Lc0 soon to support chess960 ?

Post by Hai » Wed Apr 22, 2020 8:27 am

The problem is that your comparisons is wrong.

But feel free to show me which of your different networks was stronger, than 712574 or 712576, after only the first day of training.

712576 looks stronger than 712576.

Modern Times
Posts: 2604
Joined: Thu Jun 07, 2012 9:02 pm

Re: Lc0 soon to support chess960 ?

Post by Modern Times » Wed Apr 22, 2020 9:02 am

Not wrong at all, I don't have all day and banks of machines to test every network. I ran just 712574 as suggested above by Kiudee, and 591226.

591226 will act as a baseline, and once an extensively trained network is available for chess960 we'll see what improvement there is.
.

Opinions expressed here are my own, and not necessarily those of the CCRL Group.

Hai
Posts: 548
Joined: Sun Aug 04, 2013 11:19 am

Re: Lc0 soon to support chess960 ?

Post by Hai » Sat Apr 25, 2020 8:36 pm

Good that chess 960 improve tactics.
And LC0 would avoid things like this:

[White "Stockfish 150420 64 POPCNT"] 3 cores
[Black "LC0 30x384"] (3350) 2x RTX 2080 Ti
[Result "1/2-1/2"]
[ECO "B14"]
[TimeControl "60+1"]
No tb
Game started after 1.e3


Post Reply