es gab mal eine Vorhersage-Wett-Seite (leider starb dann der CEO 50m vorm Mt.Everest Gipfel
und die Sache ging den Bach runter) , aber ein "claim" etwa "ein NN-programm gewinnt TCEC14" waere interessant.
Was wuerdet ihr wetten ? Vielleicht koennen wir ja hier solche "Aktien" handeln ? mit virtuellem
Geld natuerlich, da solche Wettboersen in Deutschland anscheinend nicht gern gesehen werden.
Ich kaufe fuer 40 und verkaufe fuer 60.
eine der Expertendiskussionen vor 3 Monaten, die ich auf die schnelle fand :
-------------------------------
https://github.com/suragnair/alpha-zero-general/tree/master/pytorch_classificationthey have ~20M parameters (1M-70M)
https://github.com/crypt3lx2k/ZerofishCurrently uses a completely different model than the one from the paper!
This model has a very different layout than the one from the paper.
Significantly reduced number of parameters in value and policy output heads.
https://www.reddit.com/r/chess/comments/7igro1/alphazero_reactions_from_top_gms_stockfish_author/[–]Harawaldr wrote:
It is also worth noting that their approach was inefficient to say the least,
because they intended to prove the general purpose quality of their method.
Whoever adapts the A0 method for chess, with the intention of creating the
best possible engine, will optimise hyper parameters for chess,
they will use tablebases, they will experiment with different state representations.
I hypothesise that the chess world will see software based on the A0 architecture,
but better, and runnable on high end consumer hardware within three years.
https://www.reddit.com/r/MachineLearning/comments/7qdwb5/is_it_possible_to_train_the_value_output_in_a/Zeta_36
picardythird wrote :
my own chess implementation of AGZ
https://github.com/trebledawson/chessmy neural network was too small, increase the number of parameters
until it was as large as my computer could store,
zeta_36 wrote: I really think DeepMind did something they didn't explain
thousands of parameters
https://sjeng.org/leela.htmlLeela Go FAQ
no new version since Oct.