The difference between Alpha Zero and chess engines is that chess engines are programmed to play chess, while Alpha Zero is programmed to learn how to play chess (or go, or any game with a defined set of rules). Meaning that the programmers have no idea how Alpha Zero does it. It ain't gonna run on your smartphone anytime soon, and it hasn't been demonstrated that it would beat Stockfish running on comparable hardware. Go and chess are just demonstrations - the true applications of Alpha Zero are said to be in areas like medical diagnosis and protein folding. IBM's Watson is being promoted for similar applications.
Everything AlphaZero
Sort:
True, do you have any questions or personal opinions.
I think that Alphazero just raised the bar when it comes chess engines at the very least. I think we'll start to see more and more improvements there, especially from established chess software companies. I also think that the way these engines play will improve human play as well. Maybe all the theory we know and the systems that've been created in regards to chess are obsolete. This is of course a very ignorant, uneducated theory. It's just me wondering really. I have, after all, only read one article about it and reviewed just one game. Sometimes a person thinking out loud can get ripped apart in these forums, so I like to keep it short