Einzelnen Beitrag anzeigen
  #12  
Alt 15.12.17, 23:05
Timm Timm ist offline
Singularität
 
Registriert seit: 26.03.2009
Ort: Weinstraße, Rheinld.Pfalz
Beitr?ge: 3.165
Standard AW: Alpha Zero schlägt Stockfish

Zitat:
Zitat von Hawkwind Beitrag anzeigen
hmm, warum schreib ich eigentlich auf Englisch?
Perhaps because the english processing part of your neural net couldn't but proceed in english after having read Ed Schröder.

Er ist ja ziemlich bissig. Die Unterstellung #4 scheint selbst unter "devil's advocate" ziemlich abwegig.

Hier noch Infos zu AlphaZero von Jemand, der ziemlich Ahnung zu haben scheint.

https://chess.stackexchange.com/ques...ding-alphazero
Zitat:
Stockfish uses alpha-beta, while AlphaZero uses Monte-Carlo. They are two very different algorithms. The alpha-beta algorithm assumes a lower/upper bound, while Monte-Carlo creates simulations from the root to leaf.
https://chess.stackexchange.com/ques...has-never-seen
Zitat:
The evaluation function of a chess engine, whether instantiated as a neural net or explicit code, is always able to assign a value to any board position. If you give it a board position, even absurd ones that would never occur in a game, it will be able to spit out a number representing how favorable it is to one player or another. Since the number of board positions in chess is unmanageably gigantic, the training can only occur on an infinitesimal sample of the game tree. The engine is not simply recalling previously calculated values of board positions, but is performing calculations based on the arrangement of the pieces. For a non-neural-net example, part of a chess engine's evaluation might be to add up the value of each piece on its side and subtract the total value of the opponent's pieces. Then, one set of parameter that would be adjusted while training would be the value of each piece.
__________________
Der Verstand schafft die Wahrheit nicht, sondern er findet sie vor - Aurelius Augustinus
Mit Zitat antworten