Re:Chess-playing neural network LC0 beats Stockfish in 100-g...
by ClassyJacket on Tuesday May 28, 2019 @07:22PM
(#20035404)
I know nothing of this system other than reading this page, but my basic understanding is that Stockfish is more of a traditional algorithm that simulates possibilities and finds the most likely path to a win ('IF I move my knight here, they could do X or Y or Z...'), but LC0 is a neural network, which don't really process absolutes, just probabilities in terms of floating point numbers. It might be something to do with that.
[ Parent
| on HN ]
Re:Chess-playing neural network LC0 beats Stockfish in 100-g...
by nikki93 on Tuesday May 28, 2019 @08:21PM
(#20035757)
Stockfish's evaluation function is; though; ultimately a floating point accumulation of such decision processes. In fact the leaf level scores are themselves floating point.
eg. check out the scores here: https://github.com/official-stockfish/Stockfish/blob/master/...
... actually realized it uses an int typedef for the 'Score' type so it's "fixed-point" I guess.
Re:Chess-playing neural network LC0 beats Stockfish in 100-g...
by ClassyJacket on Wednesday May 29, 2019 @12:51AM
(#20036944)
Sure, but it still makes sense that that number would still result in a flat zero - or be disregarded - since Stockfish is (presumably?) capable of evaluating and realizing the absolute 0 chance, no? I'm not claiming to have special knowledge of these algorithms but it makes at least intuitive sense to me.