AlphaZero Chess is not that strong ...

[CheckersGuy]

Yeah 100 elo above stockfish isn't like what they have accomplished with Go. However, I wonder how strong alphaZero could get given more training and optimizations specific to chess. Since AlphaZero was, basically, for any board game I think there might be a lot of improvements.

Let's bet the improvements will be 0.0 in 5 years' time.
This is their peak.

You can not achieve an engine much stronger than 2800 with such algorithms.

Where did you get that figure from ? Any backup ?

[Lyudmil Tsvetkov]

AlphaZero is very strong. The best proof is this whole debate that in better conditions Stockfish could be on par with it. This is not exactly definition of weak. All points about SF having not optimal setup are true but they cannot change the fact that AZ is amazingly strong chess player.

All this is not any news for people following AlphaGo development. News about AlphaGo playing on professional level broke around 2 years ago and it only got stronger since then. If anything it was surprising that they didn't tackle chess earlier. They surely tried but it only shows how good current top engines are.

Seeing what happened with AlphaGo we should not expect much more from Google at least not until they have some kind of breakthrough. They will probably publish some more games, including self-played (maybe 20-50, it's not very likely to see more). Computer chess community should start their own project to see what's possible with NN approach. Merged with domain-specific knowledge it could end up maybe even few hundred Elo stronger than AZ, who knows. It's really exciting. In Go we can already have top human level program running on ordinary PC with modern GPU and very strong amateur level running on measly laptop.

Last point: AlphaZero approach works, denying it won't change anything. Look at Leela Zero project. With much smaller network and community effort, in one month it gained around 4000 Elo above random mover from self-played games only. Now it's around 10-12 kyu level I believe. It's not very strong yet but it's amazing anyway taking into account that it still plays weird opening moves and has some major tactical blind spots. When they are corrected (with more games and more training), it will play on dan level very soon.

In what way is it strong?
Now, you have a weightlifter pulling 300 kilos from the ground, SF.
Then, you have Alpha, pulling 50. It is much weaker.
Then you add up 10 Alphas to pull the same weight, and they outperform SF.

In what way is this strong?

I don't understand, what the hell NN means.
But I bet, they don't know either.

[CheckersGuy]

NN obviously means neural network and DeepMind knows all about it. Or what do you think DeepMind is doing with their +50 phd's ?

[hgm]

Why no one ever claimed SF 8 was an incomparable break-through?

Because it did not figure out how to play like that all by itself? Because it just did what the programmers had told it to do?

[Lyudmil Tsvetkov]

But is not the NN what Alpha learned?
Why do you say 'to start with a new NN'?

The NN is a machine (a virtual one, in this case, simulated by a TPU), or actually a class of (programmable) machines, like computers. Just like there are different types of computers (PC, Mac, tablets, phones) there are different designs of NN. A NN learns what to do by training it, similar to a computer does learn to do things by explicit programming. So the knowledge leared by AlphaZero is not the NN, but the program learned by and contained inthe NN (the billions of 'weights').

What I propose is to just upgrade the NN to a better one, like you would upgrade from a smartphone to a server PC, and than train it to contain much more knowledge than the current N of AlphaZero can hold or acquire in reasonable time.

You should definitely join the Alpha team.
Was not the claim no one ever touches the code in Alpha, the only thing it knows is how to make legal moves?
Now, what would you like to change then, the rules of the game are one and the same?

[JJJ]

I m not sure Alphazero hit a wall in strenght at all. I just think it does now need a lot of time to improve because he is now in the area of high draw rate and it's gonna need a lot of training to find some hidden winning line.

But in the end, I think he could still improve and win 500 elo at least with many more time he used to train at first.

[JJJ]

Anyway, you like it or not, but Alphazero did find some nice winning lines against Stockfish.

[Lyudmil Tsvetkov]

Yeah 100 elo above stockfish isn't like what they have accomplished with Go. However, I wonder how strong alphaZero could get given more training and optimizations specific to chess. Since AlphaZero was, basically, for any board game I think there might be a lot of improvements.

Let's bet the improvements will be 0.0 in 5 years' time.
This is their peak.

You can not achieve an engine much stronger than 2800 with such algorithms.

Where did you get that figure from ? Any backup ?

Because Alpha basically plays very primitive chess, its only strength is outcalculation, and especially outbooking.
That is why it picks lines where it can outcalculate the opponent, and not knowledge-based ones.

One thing is certain: you can never achieve perfect play by just tuning patterns, unless you tune those strictly exclusively, and they certainly did not do that.

That is why SF currently is at a down: its patterns are over-tuned without exclusivity applied to them.

[Dariusz Orzechowski]

In what way is this strong?

When player X has good results playing against known strong player Y, we call player X strong.

[Lyudmil Tsvetkov]

Anyway, you like it or not, but Alphazero did find some nice winning lines against Stockfish.

I would also find those, if you give me 50 times more time than SF, or 50 times more powerful hardware.

You are all very funny: why don't you expect the same performance from SF on a mobile and SF on 100 cores?