Hacker News new | ask | show | jobs
by phoe-krk 859 days ago
From the abstract:

> We annotate each board in the dataset with action-values provided by the powerful Stockfish 16 engine, leading to roughly 15 billion data points.

So some of the learning data comes from Stockfish.

1 comments

The original comment was "for playing."

In training, traditional search is absolutely used to score positions.

In playing, search is not used. (*Except to finish out an already-won position.)