Not recent versions, all versions, and the knowledge is only the rules of chess. Exactly as AlphaZero did.
Can Leela CPU train from a PGN file instead of selfplay?
Moderators: hgm, Rebel, chrisw
-
- Posts: 3019
- Joined: Wed Mar 08, 2006 9:57 pm
- Location: Rio de Janeiro, Brazil
Re: Can Leela CPU train from a PGN file instead of selfplay?
"Tactics are the bricks and sticks that make up a game, but positional play is the architectural blueprint."
-
- Posts: 99
- Joined: Sat Mar 10, 2018 6:16 am
Re: Can Leela CPU train from a PGN file instead of selfplay?
That is what the Darkqueen net is actually. Although latest ones have added a little bit of other data as well.
-
- Posts: 99
- Joined: Sat Mar 10, 2018 6:16 am
Re: Can Leela CPU train from a PGN file instead of selfplay?
There are many different levels to do this. If you just use straight PGN it creates a very weak policy head (like Darkqueen has), because it only has data that the move played is "good", but it is not always the best one, which is ok because with a large enough sample size noise will cancel out, but all the other moves are weighted as zero, so we have no idea what other moves are important to look at.
If you throw in a few million PGNs and a few million games of selfplay data with a score for each move then you can get a very strong network like Leelenstein, and then you can go back and re-annotate those PGNs with that newly created / strongest version to get the same % probability for each move played instead of the "one-hot encoded" version, and get another jump up in strength as it completes which can take months of GPU time.
There are some other techniques you can do in between these two extremes, like policy smoothing to distribute some of the 100% training target on the move played to other moves, using simple instant rules like a distribution, or a short Stockfish search for example, which dkappe has used for some of his networks like Gyal.