Good question.
Szabo thinks that he has completely figured out AlphaZero.
That it is just a Gigantic Opening Book and nothing else.
In four hours self-training after learning basic moves, it played out all the trillions and trillions of possible chess games and so has solved chess.
I heard Szabo has been offered a Job in Deepmind as its Head Researcher.
Good question.
Szabo thinks that he has completely figured out AlphaZero.
That it is just a Gigantic Opening Book and nothing else.
In four hours self-training after learning basic moves, it played out all the trillions and trillions of possible chess games and so has solved chess.
I heard Szabo has been offered a Job in Deepmind as its Head Researcher.
If Re1 was opening book it means that the neural network tried out all other branches in this position once before and found out they lead to nothing and tried re1 as an approach that could work better then all the other branches.
What seems like a fairy tale today may be reality tomorrow.
Here we have a fairy tale of the day after tomorrow....
[quote="corres"]I think as Stockfish and other used chess engine can not make any plan
so AlphaZero also can not make it.
This is only an illusion. The effect of neural network causes this illusion.
The neural network works as a kind of huge memory. It is like a huge and very deep "opening book" or "learning file" what AlphaZero makes during the process of teaching and when it plays games.
If Stockfish would make and use learning file Stockfish would have more Elo and it would play like a plan maker program.
[/quote]
So Re1 was played why ?!
[/quote]
In the case of "classical" chess engines one could know why engine moved what it moved. But we can not see into a neural network so we can not know why it moved Re1...
But it is sure it has not any plan.
[quote="mclane"]
So Re1 was played why ?!
[/quote]
Good question.
Szabo thinks that he has completely figured out AlphaZero. :lol:
That it is just a Gigantic Opening Book and nothing else.
In four hours self-training after learning basic moves, it played out all the trillions and trillions of possible chess games and so has solved chess.
I heard Szabo has been offered a Job in Deepmind as its Head Researcher.
[/quote]
Mr.Dharan,
You have a bad style and you is an inattentive reader - that is all.
