Questions tagged [alphazero]
For questions about Deep Mind's Alpha Zero chess engine.
50
questions
43
votes
11
answers
23k
views
How is Alpha Zero "more human"?
I have a possibly naive question about AlphaZero. I have seen it described as playing in a "more human" style than other computers, but whatever it does, it gains about 100 ELO points by doing it. ...
- 8,155
36
votes
2
answers
9k
views
Understanding AlphaZero
Very recently, the news came out that Alphabet's DeepMind research team have extended their machine learning engine to play both Shogi and chess. Apparently, after only a few hours of self-learning, ...
- 3,386
33
votes
7
answers
20k
views
Hardware used in AlphaZero vs Stockfish match
I understand that AlphaZero has to use a different kind of hardware than regular Stockfish. I would expect that the hardware has a large effect on engine strength. That's why I wonder whether there ...
- 20.7k
19
votes
5
answers
7k
views
Was AlphaZero taught castling?
Since castling involves one player moving two pieces in one move, was AlphaZero just given this move, in the same way it was told that players take turns and a win is better than a loss? In other ...
- 215
17
votes
7
answers
8k
views
Did AlphaZero also have to learn that each piece has a value?
When AlphaZero learned on its own that a queen has a value of 9, a pawn 1, knight 3, etc., then did it also need to learn on its own that a piece has a value? Or was it that the concept of value was ...
- 415
16
votes
6
answers
10k
views
Is it possible Alpha Zero will eventually solve chess?
I googled this question and ended up here. From previous questions, it seems Alpha Zero has not solved chess yet and if it does it will take a very long time. Nonetheless, it may be possible one day. ...
- 193
13
votes
1
answer
11k
views
Which is better-Stockfish 10 or AlphaZero?
I know AlphaZero (an AI) is one of the best top chess engines but Stockfish is also improving. Which one is better at chess?
- 241
12
votes
3
answers
19k
views
Where can you play against AlphaZero?
I have read a few articles about the game between Stockfish and AlphaZero. But is there a place where anyone can play against AlphaZero?
- 873
12
votes
2
answers
2k
views
How does AlphaZero learn to evaluate a position it has never seen?
Following up from answers to:
Understanding AlphaZero
My question would be how the neural net "learns" what to do in a position it hasn't encountered. Saying the actual AZ executes an MCTS using the ...
- 137
10
votes
3
answers
3k
views
Did Stockfish have access to an opening book in its games against AlphaZero?
I have seen some commentary that implies that Stockfish did not have access to an opening book in its much-reported match with AlphaZero, but I cannot find a conclusive citation, and my inability to ...
- 347
9
votes
2
answers
26k
views
List of how AlphaZero evaluates openings
In his process of learning Chess and getting stronger and stronger each second AlphaZero learned openings. Some were played often at the beginning, but lost in AlphaZero's popularity over the time (e....
- 399
9
votes
5
answers
2k
views
Level the field when letting computers play against each other
I just read this question and it made me wonder: when setting up such a game/comparison - shouldn't you try to level the computing capacity field? Or at least limit how much energy each "player" is ...
- 259
9
votes
1
answer
2k
views
Is AlphaZero open source?
There are a lot of fake AlphaZero projects on GitHub. While I was googling if it was open-source, the answer in most places is yes, but I couldn’t find it anywhere. Is there a link to where the code ...
- 873
9
votes
3
answers
2k
views
Knowing that AlphaZero beat Stockfish 8 (28 wins, 0 losses, and 72 draws), what would AlphaZero's rating be?
I found three different questions here on SEC that asked what Stockfish 8's rating was for the match against AlphaZero, but none that asked what AlphaZero's rating should have been after (if it had ...
- 32.4k
8
votes
3
answers
1k
views
Lecture/Book on AlphaGo/AlphaZero
I am very interested in how AlphaGo resp. AlphaZero works. It seems to me, the related Google Papers are very dense and not easy to read.
Is there any textbook or lecture that explains on a technical ...
- 539
7
votes
2
answers
2k
views
Is there a reason AlphaZero hasn't been allowed to learn for longer?
Why don't the developers of AlphaZero let it learn for days, or even months on end? Wouldn't its neural network become much stronger, at the cost of just occupying some computational resources for a ...
- 19.8k
7
votes
2
answers
1k
views
What was the source of randomness in the Alpha Zero vs Stockfish match?
My understanding is that Alpha Zero's algorithm was fixed after the "learning" phase, so would it come up with the same move for every position? Was the variability in the match entirely due to a ...
- 5,757
7
votes
1
answer
1k
views
How does Alphazero Respond to the QGD?
I recently started reading Game Changer, the new(ish) book on Alphazero, and I'm really enjoying it. My one issue so far is that I can't find, in the book or elsewhere, games where Alphazero played ...
- 73
6
votes
2
answers
4k
views
Alpha Zero vs Lc0 - time for self-play
Alpha Zero defeated Stockfish after only 9 hours of self-play. On the other hand, Lc0, has been training for several months and while attaining impressive strength (to say the least...) it's still not ...
- 1,268
6
votes
1
answer
420
views
AlphaZero’s search procedure
I am aware of related questions and great answers on the same topic such as Understanding AlphaZero. My questions are related to the following figure on AlphaZero's search procedure
This figure ...
- 3,525
5
votes
4
answers
2k
views
Book recommendation for chess engine building
This might be too broad a question, but can anyone recommend specific books that deal with chess engines, including hard-coded ones such as Stockfish, Komodo, and other ones also based on neural ...
- 256
5
votes
3
answers
1k
views
Where is the neural network used in MCTS?
Where does an engine (such as AlphaZero) use its neural network in MCTS? It shouldn't be needed to evaluate positions reached via MCTS, since I'm under the impression that only terminal game states ...
- 19.8k
5
votes
1
answer
281
views
Combining AlphaZero methods with Stockfish methods [closed]
NOTE: I asked a similar question at the ai stackexchange, but didn't get any satisfying answers, so I will post an updated version of the question here, to see if someone with more of a chess ...
- 649
5
votes
2
answers
562
views
Does Google's new quantum computer mean we will soon have a more powerful AlphaZero?
According to a paper published in Nature Google has developed a quantum computer.
Abstract
The promise of quantum computers is that certain computational tasks
might be executed ...
- 94.5k
5
votes
2
answers
959
views
Is AlphaZero deterministic?
AlphaZero searches for moves using Monte Carlo Tree Search (MCTS). As I understand MCTS, it takes the root position, goes to a child node, and then plays random moves until one side wins. Since the ...
- 26.1k
4
votes
1
answer
610
views
Is there any analysis of the Carlsen-Caruana World Championsip match by AlphaZero?
How cool would it be to have the world's most powerful computer analyzing games played in a match between the world's two strongest human players? Has it happened yet and is the analysis available?
- 94.5k
4
votes
3
answers
1k
views
Why does AlphaZero evaluate on the order of 1000 times less moves than Stockfish?
According to Deep mind's article, Alpha Zero searches roughly 10,000 moves per decision, while Stockfish is around 10,000,000 moves. I'm aware that Alpha Zero uses Monte Carlo Tree Search while ...
- 19.8k
3
votes
2
answers
239
views
Alphazero vs Stockfish
Is the particular ending really a draw? Someone may try will FinalGen , please?
[Title "White to move"]
[fen "8/1B2k3/p7/P6p/1n1p2pP/6P1/4KP2/8 w - - 0 1"]
- 259
3
votes
1
answer
325
views
What kind of hand-crafted features do top performing family of engines use to evaluate a position?
AlphaZero paper which introduced the engine which allegidly outperformed stockfish mentioned the following:
These programs evaluate positions using features handcrafted by human
grandmasters and ...
- 4,738
3
votes
1
answer
4k
views
What is Alpha Zero's Play Style? (Including its Openings and its Middlegame) [closed]
If you want to read about its opening strategy, it is down below the link. If you want to read about its middlegame strategy or basic play style, it is at the top
I want to know how I can play more ...
- 31
3
votes
1
answer
180
views
Guidance on a new idea for an engine design?
for my A-level computing project I was going to do an investigation into a type of engine design which incorporates Neural Networks that I haven't seen elsewhere; as part of this project I need to ...
- 31
3
votes
1
answer
2k
views
What are the strategies of Alphazero? [closed]
Alphazero is an amazing chess engine, one of the best in the world. It also has a more "human" playing style than other engines. Furthermore, understanding this engine might be vital to improving the ...
- 41
3
votes
0
answers
355
views
Understanding Leela Chess Zero (Lc0) "verbose-move-stats output"
I'm trying to fully understand the Leela's verbose-move-stats output:
Are the statements below correct?
The move that will be played is e2e4, since it's the one with the greatest count of visits (N: ...
- 168
2
votes
3
answers
4k
views
The games AlphaZero lost
AlphaZero lost 6 games to Stockfish in a 1000-game match from the starting position, at least one of which was with the white pieces: A game A0 lost with white. For some reason, these games haven't ...
- 465
2
votes
3
answers
2k
views
To what extent (if at all) is the new Komodo 13 MCTS an AlphaZero Lite?
The new Komodo 13 MCTS has been announced with the claim that:
Komodo 13 MCTS thinks like no other chess engine in the world.
Inspired by the ideas of AlphaZero and Leela, Larry and Mark set out
...
- 94.5k
2
votes
3
answers
12k
views
What is the Elo rating of Stockfish version that played AlphaZero?
AlphaZero has defeated Stockfish with all novelties and brilliant strategic sacrifices. But some people were disputing its win, because Stockfish was running on weak hardware (1 GB RAM), not access to ...
- 145
2
votes
1
answer
590
views
Availability of AlphaZero and A. Grischuk's idea in the Anti-Moscow Gambit vs L. Ding
I was watching some Fide Candidates post play conference and Grischuk mentioned that a certain move was suggested by AlphaZero. So I was googling a little and have a stupid question. Where is ...
- 163
2
votes
1
answer
338
views
What was it that AlphaZero did that dazzled experts?
Wired magazine recently said
Most remarkably, AlphaZero uncovered new approaches to the game that
dazzled chess experts.
Just what was unique about alpha zero that was so dazzling?
- 3,871
2
votes
2
answers
450
views
AlphaZero and the revenge of the Artificial Intelligence
Answering this question, @SmallChess offered this answer:
Before Google's chess journey, the chess engine community believed:
AI chess will not beat classical programming because neural networks
...
- 3,525
2
votes
2
answers
838
views
Alphazero-Stockfish game 10 move 19
Given sufficient analysis time, say 24h per move on an decent computer, can current top chess engines (including Stockfish) eventually "see" that move 19 by Alphazero is actually good, and even ...
- 41
2
votes
1
answer
480
views
Best way to loop through all possible moves (Alpha Beta)?
Essentially I'm just trying to create a somewhat efficient chess ai using an Alpha Beta Pruning approach for a future neural network to train against. I know I could use an existing program but this ...
2
votes
1
answer
257
views
Why does AlphaZero implement time steps in its input layers?
I'm trying to build my own chess engine using neural nets, and have been researching papers from Deepmind about their algorithm implemented in AlphaZero. Since chess is a game with perfect information ...
- 21
2
votes
1
answer
588
views
What was one shortcoming of alpha zero that Kasparov noted?
I [Kasparov] can look at AlphaZero’s games and understand the
potential weaknesses. And I believe it has made some inaccurate
evaluations, which is natural.
The reason he says that is AlphaZero ...
- 3,871
2
votes
1
answer
318
views
Is there any game where AlphaZero wins against Stockfish thanks to a knight pair?
I have seen some of the games that were published in the match between AlphaZero and Stockfish.
In many games, as when it beats Stockfish as black playing the Berlin Defense, AlphaZero appears to ...
user18196
2
votes
1
answer
266
views
Amount of search per decision: GMs versus AlphaZero and Stockfish [duplicate]
In Demis Hassabis' presentation at the the Kinds of Intelligence Symposium in January 2018, grandmasters are compared to AlphaZero and Stockfish along an interesting metric: how many moves are ...
- 3,525
1
vote
2
answers
96
views
Does a reinforcement learning style model actually need to be combined with a search algo to produce the best moves?
I have recently been studying up on Machine Learning based chess engines and have begun to develop one of my own. I was wondering, realistically, doesn't the board technically contain all of the data ...
- 19
1
vote
1
answer
282
views
Alphazero Learning vs Playing mode
Is there any fundamental difference (in algorithms etc.) between Alphazero training by playing against itself and when playing a match against Stockfish?
Or were the match games basically treated as ...
- 20.7k
-1
votes
1
answer
426
views
How is AlphaZero defferent from Stockfish?
Obviously one is better than the other and AlphaZero uses machine learning and all, but what is the real reason for their different style of play? Given a position, both evaluate the next best move ...
-1
votes
2
answers
2k
views
Are AlphaZero's openings objectively better than the others? [closed]
Are AlphaZero's openings objectively better than the others? If so, why haven't top grandmasters like Carlsen exclusively adopted them in their own games?
- 465
-2
votes
2
answers
261
views
Why is it seemingly impossible to follow high level chess games?
According to this, AlphaZero was a really good chess engine that learned through playing against itself, and it then it never lost against any other chess programs. I think it probably also said it ...
- 410