Re:Human-like neural network chess engine trained on lichess...
by gamegoblin on Sunday January 17, 2021 @04:15PM
(#25815028)
Is there a way to treat resignation as a "move"? Even though the win probability is zero by definition, it still may be the most accurate move prediction in certain scenarios.
[ Parent
| on HN ]
Re:Human-like neural network chess engine trained on lichess...
by reidmcy on Sunday January 17, 2021 @04:31PM
(#25815165)
Yes, the output is just a large vector with each dimension mapping to a move. But we'd probably do it as different "head" so have a small set of layers that are trained just to predict resignations. Both options would break the lc0 chess engine we use for stuff like the Lichess bots though.