Umelá inteligencia Google sa naučila za 4 hodiny sama hrať šach lepšie ako najlepší ľudia a softvér
Diskusia k článku: Umelá inteligencia Google sa naučila za 4 hodiny sama hrať šach lepšie ako najlepší ľudia a softvér
Prispievajte do diskusií ako
prihlásený užívateľ.
Komentár, na ktorý odpovedáte:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Re: Stockfish
Od: ttt123
|
Pridané:
2017-12-07 17:12:52
single netw:
In AlphaGo Zero, self-play games were generated by the best player from all previous iterations. After each iteration of training, the performance of the new player was measured against the best player; if it won by a margin of 55% then it replaced the best player and self-play games were subsequently generated by this new player. In contrast, AlphaZero simply maintains a single neural network that is updated continually, rather than waiting for an iteration to complete.
|