Like Morphy and Fischer Alpha Zero is done.

This from a guy that can't figure out the quote function?

1. I think the creators of Stockfish have the right to issue a challenge and AlphaZero should accept. Both bring whatever computing power they want or accept the same as the opponent is using, their choice. No CPU blaming after the fact. I hope this happens.

2. Some other programs have started to use similar techniques to AlphaZero and seen immediate increases in ELO. This is fundamentally a different way for computers to play the game and there might be a rock/paper/scissors analogy, but for now it's rock and paper.

this is a hit and run of dubious importance.  if alpha zer0 doesn't compete at the computer chess championship, then stockfish reigns supreme.

That's interesting about other programs, have you got any links relating to that?

AlphaZero was just a test run for the military. In the near future, armed drones will fly around a battlefield, controlled like chess pieces by AI.

Or driverless cars that use the element of surprise to try to get you to make a mistake when doing the school run

I'm looking for the reference, but I think it was in the comments of a YouTube video. Sigh. I'll keep looking for the ELO rise, but it was one of the top 20 programs, not one of the top 5.

Have you heard about the rematch Alpha Zero vs Stockfish, Komodo, & Houdini? 

I could only find this:

https://groups.google.com/forum/#!topic/fishcooking/ExSnY8xy7sY

Basically an open challenge from one of Stockfish's devs asking for favorable setting for Stockfish. I hope we get to see something like that happen, but other analysis suggest that so many games were draws that winning in Chess at that level of play is not only hard, but rare. They say the winner is chess is the one that makes the second to last blunder... in this case I think it only takes one.

 If I understand correctly, there is a claim there that Stockfish 9 beats Stockfish 8 almost as clearly as AlphaZero did!

I read this claim but was unable to find a citation from a reliable source. At this point I find it to be rumor or trolling.

I don't believe an opening book would give Stockfish 50 points against AlphaZero, which looks rock solid in the opening. Indeed in the match, Stockfish was also solid in the openings and went wrong in the middlegames.

I'm  not sure I agree with this. Here's my reasoning. The opening books represent the basis for solid middle-games. That means they set up situations for optimal positioning and movement of your pieces in the middle game and defend against your opponent doing the same.

If Stockfish's middle game was to blame for its loses, then we can clearly see that this might be from a lack of foresight. See game 10 for an example... and you start to wonder if Stockfish would have gotten to that middle game with trapped pieces if it had an opening book to give it a way out.

Note that I'm not saying Stockfish would have won that game or even drawn it with an opening book. I'm only suggesting that the game might have played differently if Stockfish had used an opening book that optimized pawn placement and allowed better movement for its minor pieces.

Opening books are made by men. Wouldn't that be man & machine vs machine.

There is some truth in that, but the problem is that opening books are less and less effective for even very powerful engines, never mind AlphaZero. All they can really do is avoid some disasters and defer the real battle. The very best an opening book can ever do is to use previous games as samples to use in analysis (which relies on trusting that the games are adequately accurate). For example, this can go badly wrong if you are playing the French against AlphaZero, even if it might be ok against a GM!

I disagree. Opening books can be invaluable because they are based on years of experience that the computer doesn't have to do itself using up plys, CPU cycles and game time.

Think of it as nearly infinite plys by humans codified into opening moves. Again, in the relatively few games that AlphaZero won (the vast majority were draws), it was because Stockfish effectively blocked itself in with its opening.

Yeah, but "infinite ply by humans" (you don't really mean that) is as much use to Stockfish as the games of average club players are to Carlsen. Not very much.

Openings are one of the few things in Chess that you can really study, as a human, to any depth. The other is endgame. Middle game is obviously where computers shine because they can look at so many more options per second than a human can and divine favorable solutions.

Opening theory, and all that it has to offer, from the 20 possible first moves and 20 possible replies, out to six or seven variations have been looked at for decades by humans and that information, for a computer, in valuable because it's time the computer doesn't have to waste.

Simple thought experiment: Do you think Stockfish (on the system that DeepMind ran it on) was capable of creating a solid opening equal to that of an opening book?

The question is, of course, rhetorical, because no, it can't. That's why it get's a bump in ELO from using the book. It ends up in more favorable positions when it uses an opening book because these books are built with decades of game knowledge and positional theory built in...

So again, there may be something to the theory that Stockfish would have done radically better with an opening book.

My view is that by human standards it could generate a good opening book (AlphaZero would build a significantly better one!). The body of very strong computer games might be used to generate a better book. Of course to do so, they need to be given free rein some time in the opening!