tsoj

6 ms·

1.
by tsoj 7d ago
It's not really what you're looking for, but I made this lichess BOT that makes stupid (but relevant) comments on your moves from time to time: https://lichess.org/@/Annie_Archy/all
2.
by tsoj 7d ago
The next closest is probably Torch currently
3.
by tsoj 7d ago
Join the Stockfish discord server! There is a very active community who builds their own chess engines there. (If you ever end up going for it, don't use AI it takes away a lot of the fun)
4.
by tsoj 7d ago
LMAO
5.
by tsoj 7d ago
Maths is "turn based" albeit it has practically infinite available moves per turn. But I think the latter matters less than it may appear at first: At least in practice there isn't much different (for both humans and computer
6.
by tsoj 7d ago
Usability for human analysis will likely be more successful going a similar direction Maia is going with the lc0 like net trained to make human moves. SF is inherently so far removed from the human approach to chess (relatively speaking), t
7.
How would LLMs vote in upcoming German state elections?(tsoj.link)
1 points by tsoj 11d ago|1 comments
8.
by tsoj 11d ago
The Wahl-O-Mat ( https://wahl-o-mat.de/ ) is a quiz-based website that asks user to select it they agree or disagree with various statements and then matches how similarly parties running in elections answer compared to the u
9.
by tsoj 1mo ago
what is this swindle
10.
by tsoj 1mo ago
wtf the emoji didn't show up
11.
by tsoj 1mo ago
?
12.
by tsoj 8mo ago
I don't work on Stockfish, but I can suggest using ShashChess instead. It is Stockfosh, but on top of that, it has been improved to capture the spirit of the human ingenuity and creativity.
13.
by tsoj 11mo ago
The TRM paper addresses this blog post. I don't think you need to read the HRM analysis very carefully, the TRM has the advantage of being disentangled compared to the HRM, making ablations easier. I think the real value of the arcpriz
14.
by tsoj 2y ago
Yeah, humans are very similar. We have intuitive immediate-next-step-suggestions, and then we apply these intuitive next steps, until we find that it lead to a dead end, and then we backtrack. I always say, the way we used LLMs (so far) is
15.
by tsoj 2y ago
This is NOT the paper, but probably a very similar solution: https://arxiv.org/abs/2009.03393
16.
by tsoj 4y ago
The thing is, Sutter's approach is much more sensible when looking at real world adaptation. You can start using the cppfront transpiler, and importantly, if it doesn't work out, you can just use the C++ code generated by cppfront
17.
by tsoj 4y ago
And for the stuff that matters, it is usually possible to tinker around enough to get it pretty much as good as Rust or C++.
18.
by tsoj 5y ago
> no language with GC gets closer than a 2-3x the time of a non-GC language This is wrong. At least, it is not true generally. I wrote a 1:1 port of my C++ chess engine, and after a bit of fiddling with compiler options the Nim version w