←back to thread

Grandmaster-level chess without search

(github.com)
365 points lawrenceyan | 4 comments | 17 Oct 24 19:13 UTC | HN request time: 0.656s | source
1. osti ◴[17 Oct 24 23:05 UTC] No.41874706[source]▶
https://lczero.org/blog/2024/02/how-well-do-lc0-networks-com...

The best neural network chess engine's authors wrote about this deepminds publication.

replies(1): >>41878014 #
2. stabbles ◴[18 Oct 24 10:30 UTC] No.41878014[source]▶
LC0 hasn't been the best neural network chess engine since Stockfish added NNUE in 2020.
replies(2): >>41880509 #>>41881273 #
3. vlovich123 ◴[18 Oct 24 15:47 UTC] No.41880509[source]▶
That isn’t quite accurate. Stockfish’s NNUE was trained against Leela evaluated positions.

> Generally considered to be the strongest GPU engine, it continues to provide open data which is essential for training our NNUE networks. They released version 0.31.1 of their engine a few weeks ago, check it out!

The main difference is that Stockfish is targeting to run on the CPU while Leela targets the GPU. That stockfish is able to be competitive with Leela is of course impressive.

https://lichess.org/@/StockfishNews/blog/stockfish-17-is-her...

4. osti ◴[18 Oct 24 16:58 UTC] No.41881273[source]▶
True, but since stockfish uses a way way smaller network, I still prefer to think of stockfish as the traditional engine.