53
votes
Was AlphaZero taught castling?
Other answers are right in that AlphaZero was implicitly told that castling is a move that exists, I'll elaborate on how exactly that happens by explaining the inputs and outputs of the neural network....
- 1,004
50
votes
Was AlphaZero taught castling?
was AlphaZero just given this move
It was given this move in the same way that it was given all other moves. It started with complete knowledge of the rules of the game. In machine learning terms, ...
- 7,504
48
votes
Is it possible Alpha Zero will eventually solve chess?
No, it can't.
Even if you had it train on a zillion self-playing games and it miraculously achieved perfect play somehow, we would have no way of proving that it had achieved perfect play without ...
- 27.7k
37
votes
How is Alpha Zero "more human"?
Page 5 in the paper has your answer:
... AlphaZero compensates for the lower number of evaluations by using its deep neural network to costs much more selectively on the most promising variations - ...
- 22.4k
30
votes
Accepted
Where can you play against AlphaZero?
As already explained by other answers, AlphaZero itself is not publicly available.
The community has created Leela Chess Zero, a chess engine that started by using the same basic concepts as AlphaZero,...
- 1,004
28
votes
Did AlphaZero also have to learn that each piece has a value?
It sort of did. "Sort of" because once you examine how neural networks work it's not clear what AlphaZero is actually learning.
AlphaZero has a neural network evaluation function. How that ...
- 26.1k
25
votes
Accepted
Hardware used in AlphaZero vs Stockfish match
That's why I wonder whether there have been any attempts made to
provide comparable hardware to both.
This is Google you're talking about! So the answer is obviously "No".
From the original paper ...
- 94.5k
24
votes
Accepted
Understanding AlphaZero
How does AlphaZero select a move in the search?
This is very obvious from the paper.
Each simulation proceeds by selecting in each state s a move a with low visit count, high move probability and ...
- 22.4k
21
votes
Accepted
Which is better-Stockfish 10 or AlphaZero?
We can't say for sure since AlphaZero is a private engine, i.e. we don't have games between it and the latest versions of Stockfish. Still, if AlphaZero hasn't improved since it was unveiled, it will ...
- 26.1k
19
votes
How is Alpha Zero "more human"?
Most strong engines emphasize looking very deeply, at the expense of having a superficial evaluation function. In the AlphaZero paper, they say that Stockfish looks at 70 million positions per second.
...
- 27.7k
16
votes
Did AlphaZero also have to learn that each piece has a value?
The short answer is: No.
AlphaZero learns to evaluate only the position. The position consists of all the pieces and their placement on the board (plus castling and en-passant information). There is ...
- 271
15
votes
Hardware used in AlphaZero vs Stockfish match
I think it's best if I elaborate on your second point with an example move in the game 1 between AlphaZero and Stockfish which also served to satisfy my curiosity today.
the time limit of 1 min/...
- 151
13
votes
How is Alpha Zero "more human"?
AlphaZero already seems to play like a regular "centaur" -> correspodence GM with an engine assistance.
As an FM I'd get much more enjoyment of playing AlphaZero vs a regular engine.
One comparison ...
- 1,009
13
votes
Where can you play against AlphaZero?
No, you can't play against AlphaZero, unless you get a special access from DeepMind.
- 1,898
12
votes
The games AlphaZero lost
AlphaZero hasn't solved chess, and it won't until it can find a strategy that results in the same outcome of the game given perfect play from both sides. According to Wikipedia:
A solved game is a ...
- 3,022
12
votes
Is there a reason AlphaZero hasn't been allowed to learn for longer?
Check out the AlphaZero paper, figure 1.
As you can see from the leftmost figure, you can let AlphaZero train, but that doesn't mean it'll improve. In fact, AlphaZero peaked at a level slightly above ...
- 26.1k
12
votes
Was AlphaZero taught castling?
The AlphaZero paper "Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm" by David Silver et al. contains this explanation
Knowledge of the rules is also ...
- 221
11
votes
Accepted
List of how AlphaZero evaluates openings
You can find the complete table in their paper. See table 2 in the arXiv version linked below:
Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning Algorithm https://arxiv.org/...
- 12k
11
votes
Alpha Zero vs Lc0 - time for self-play
In short, because DeepMind has TPUs. Lots and lots of TPUs.
By far, the most time-consuming part of the training process is generating self-play games. DeepMind used 5000 TPUs to generate self-play ...
- 211
11
votes
Accepted
10
votes
Accepted
Lecture/Book on AlphaGo/AlphaZero
EDIT
@unutbu's link in the comment is a good introductory read.
Solid understanding for AlphaZero most likely require a quantitative degree (PhD?). Are you asking for a crash course in AlphaZero?
...
- 22.4k
9
votes
How is Alpha Zero "more human"?
It is as easy to jump on a bandwagon saying Alpha-Zero's play is 'more' human than previous computer chess programs as it is to jump on the opposite wagon and say Alpha-Zero's play is entirely 'alien'....
- 1,255
9
votes
Accepted
Did Stockfish have access to an opening book in its games against AlphaZero?
Both Nakamura and the StackExchange answer were correct - Stockfish didn't use an opening book. I think Google did that to make a fair comparison as AlphaZero wasn't using an opening book as well.
...
- 22.4k
9
votes
Accepted
Does Google's new quantum computer mean we will soon have a more powerful AlphaZero?
TLDR: No
Reasoning: Quantum Computers in the short to medium term only have some very specific things efficiently (factoring and discrete log and a few others). The algorithm google showed "quantum ...
- 1,088
9
votes
Did AlphaZero also have to learn that each piece has a value?
One central piece of alphazero is the neural evaluation function. This function takes the board as input and produces a value as output. The only input is what pieces are on the board and where.
So, ...
- 191
8
votes
Hardware used in AlphaZero vs Stockfish match
One of the original authors of Stockfish answers the specific complaints you mentioned here:
Meanwhile Chess.com also received a lengthy comment from one of the
original Stockfish authors, Tord ...
- 305
8
votes
How does AlphaZero learn to evaluate a position it has never seen?
The evaluation function of a chess engine, whether instantiated as a neural net or explicit code, is always able to assign a value to any board position. If you give it a board position, even absurd ...
- 235
8
votes
Knowing that AlphaZero beat Stockfish 8 (28 wins, 0 losses, and 72 draws), what would AlphaZero's rating be?
Here's an online calculator for this. 28 wins, 0 losses, and 72 draws is an elo difference of exactly 100.
Do remember that this score +28 =72 -0 score was against a handicapped version of Stockfish....
- 26.1k
8
votes
Is it possible Alpha Zero will eventually solve chess?
Nope. AlphaZero's entire architecture (and the architectures of most other engines) is such that it cannot "find a solution", and in any case, if a solution was to be found it would be ...
- 1,858
Only top scored, non community-wiki answers of a minimum length are eligible