TalkChess.com

dkappe wrote: ↑

Thu Jan 23, 2020 5:01 am

carldaman wrote: ↑

Thu Jan 23, 2020 2:23 am

Hi Dietrich,

Where can we get this Medium Ender net that works with Scorpio?

Thanks!

It’s downloadable from the Scorpio github. I think it’s in the Maddex zip file.

OK, thanks, I'll check it out again.

Hi Dietrich,

Where can we get this Medium Ender net that works with Scorpio?

Thanks!

AFAIK, FatFritz (FF) is also trained on human and engine games as well, not just adopting a 'zero' approach. The trouble is that it becomes an expensive program to run, requiring a high-end RTX nVidia graphics card [for optimum performance]. Chessbase would do well to add a learning feature for its ...

The MCTS version has trouble winning simple endgames, Android or not, from what I've seen.

Werner wrote: ↑

Sun Jan 19, 2020 12:40 pm

391 Rodent IV 0.22 x64 1CPU: 2896 - 900 games (+9 to v. 0.21)

Werner Schüle
CEGT-Team

Is this the Rodent IV released on Nov 17? I wasn't aware of the different versions, 0.21 and 0.22.

Thanks!

Everyone is obsessed with Elo, yet hardly anyone understands how important engine learning is in this regard. :shock: I suspect this is THE reason Learning died. The problem with it is that an engine with Learning will have a lower Elo on games. What seems to happen is that the engine faces a posit...

Yes, it's very hard to make up a difference of several hundred Elo points, even with learning enabled and no book used.

@Robert Flesher I have to run FF on a slow GPU, so it's weaker tactically, and a learning engine can latch on to such weakness and play lines FF does not like. Your fast GPU FF is much stronger and far less susceptible to this effect against a lesser engine. The principle still holds, since even the...

I don't know about your version of Leela, but my version (I also use Fat Fritz, and SF if necessary) uses the same tried-and-true approach in the Mar del Plata (and with better eval): attack the Kingside while the opponent thinks he's winning on the Q-side. Of course, everything is open to debate, a...

But, hey, Jeremy's Bernstein open implementation of learning for Stockfish from 2014 is still here: https://open-chess.org/download/file.php?id=597&sid=bdcf36b14b1783b942d39c6bac1adc60 I still don't get why nobody has implemented it for latest Stockfish and given an up to date engine public learnin...