Questions tagged [chess]

For questions about chess playing engines such as AlphaZero and Leela, and questions about AI methods and theory related to Chess. (Questions purely about chess are off-topic but can be asked on SE:Chess.)

Filter by
Sorted by
Tagged with
3 votes
1 answer
198 views

How does learning the moves of chess show up in a neural network?

Is learning the moves a special case or just the same sort of thing that happens as the AI learns strategy? If you take two different neural networks and teach them each how the pieces move, what ...
user avatar
  • 141
18 votes
2 answers
3k views

How do neural networks play chess?

I have been spending a few days trying to wrap my head around how and why neural networks are used to play chess. Although I know very little about how the game of chess works, I can understand the ...
user avatar
2 votes
1 answer
83 views

Can AlphaZero develop significantly different playing styles (depending on the random games from which it learrns)?

There is a quite popular video analysing a chess game AlphaZero vs. AlphaZero, called "the perfect game". It leaves some questions open and I'd like to ask them here: Did the two copies of ...
user avatar
1 vote
0 answers
42 views

Examples of rationalizable AI

The marvelous book Game Changer: AlphaZero's Groundbreaking Chess Strategies and the Promise of AI gave rise to this question. It is - in my opinion - a perfect example of rationalizing a piece of AI ...
user avatar
2 votes
1 answer
63 views

Reason why chess neural network might not be training

I've been trying to use a Stockfish-like chess evaluation neural network for the past few weeks but to no avail. I wanted to get some other opinions about why my current methods haven't worked. Input: ...
user avatar
  • 21
3 votes
2 answers
319 views

Why does Alpha Zero's Neural Network flip the board to be oriented towards the current player?

While reading the AlphaZero paper in preparation to code my own RL algorithm to play Chess decently well, I saw that the "The board is oriented to the perspective of the current player." I ...
user avatar
3 votes
1 answer
466 views

How does the Alpha Zero's move encoding work?

I am a beginner in AI. I'm trying to train a multi-agent RL algorithm to play chess. One issue that I ran into was representing the action space (legal moves/or honestly just moves in general) ...
user avatar
5 votes
1 answer
200 views

Clarifying representation of Neural Nerwork input for Chess Alpha Zero

In the Alpha Zero paper (https://arxiv.org/pdf/1712.01815.pdf) page 13, the input for the NN is described. In the beggining of the page, the authors state that: "The input to the Neural Network ...
user avatar
  • 63
0 votes
2 answers
71 views

What is the meaning of the terms in this evaluation function for chess?

I'm trying to improve my evaluation and I saw this here ...
user avatar
5 votes
1 answer
159 views

Do AlphaZero/MuZero learn faster in terms of number of games played than humans?

I don't know much about AI and am just curious. From what I read, AlphaZero/MuZero outperform any human chess player after a few hours of training. I have no idea how many chess games a very talented ...
user avatar
  • 153
0 votes
0 answers
290 views

How to implement very simple move-ordering for alpha-beta pruning

I've done implementing alpha-beta, and transpositional table on my search tree algorithm so I decided to implement move-ordering next. But once I implemented it, it's way more longer to respond than ...
user avatar
3 votes
0 answers
194 views

Stack of Planes as the Action Space Representation for AlphaZero (Chess)

I have a question regarding the action space of the policy network used in AlphaZero. From the paper: We represent the policy π(a|s) by a 8 × 8 × 73 stack of planes encoding a probability ...
user avatar
  • 107
2 votes
1 answer
81 views

In AlphaZero, do we need to store the data of terminal states?

I have a question about the training data used during the update/back-propagation step of the neural network in AlphaZero. From the paper: The data for each time-step $t$ is stored as ($s_t, \pi_t, ...
user avatar
  • 107
2 votes
2 answers
88 views

What happens when an opponent a neural network is playing with does not obey the rules of the game (i.e. cheats)?

For example, if AlphaZero plays with an opponent who has a right to move chess figures any way she wants, or make more than 1 move in a turn? Will a neural network adapt to that, as it adapted to an ...
user avatar
  • 129
1 vote
0 answers
143 views

To solve chess with deep RL and MCTS, how should I represent the input (the state) to a neural network?

I'm wanting to build a NN that can create a policy for each possible state. I want to combine this with MCTS to eliminate randomness so when expansion occurs, I can get the probability of the move to ...
user avatar
0 votes
1 answer
57 views

Were AI strategies identified at go or starcraft games and how?

When an AI is trained to play an opposing game, such as chess or go, it can become very strong. I have read in an article (non-scientific) the claim that AI strategies were identified by scientists ...
user avatar
3 votes
0 answers
60 views

Where does reinforcement learning actually show up in Deepmind's game engines?

From the brief research I've done on the topic, it appears that the way Deepmind's Alphazero or Muzero makes decisions is through Monte Carlo tree searches, where in the randomized simulations allows ...
user avatar
5 votes
1 answer
268 views

What are some resources for coding some artificial intelligence techniques in the context of games?

I know the most basic rudimentary theory on AI, and I want to delve into actual practical coding with AI and machine learning. I already know a decent bit of coding in C++ and I'm learning Python ...
user avatar
  • 53
0 votes
0 answers
37 views

How does (or should) AlphaGoZero (which does chess) fare against Deep Blue?

Deep blue is good at chess, but is more "hand-coded" or "top-down". https://en.wikipedia.org/wiki/Deep_Blue_(chess_computer) AlphaGoZero is "self-taught", and at Go is very much super-human. https://...
user avatar
5 votes
1 answer
1k views

How powerful is OpenAI's Gym and Universe in board games area? [closed]

I'm a big fan of computer board games and would like to make Python chess/go/shogi/mancala programs. Having heard of reinforcement learning, I decided to look at OpenAI Gym. But first of all, I would ...
user avatar
  • 63
3 votes
0 answers
477 views

How did MuZero learn the rules of chess?

Google says that their new AI program MuZero learnt the rules of chess and some other board games without being told so. How is this even possible? https://towardsdatascience.com/deepmind-unveils-...
user avatar
12 votes
1 answer
6k views

Are humans superior to machines in chess?

A friend of mine, who is an International Master at chess, told me that humans were superior to machines provided you didn't impose the time constraints that exist in competitive chess (40 moves in 2 ...
user avatar
  • 221
1 vote
1 answer
69 views

Alpha Zero queen promotion

"The final 9 planes encode possible underpromotions for pawn moves or captures in two possible diagonals, to knight, bishop or rook respectively. Other pawn moves or captures from the ...
user avatar
1 vote
1 answer
135 views

How to deal with invalid output in a policy network? [duplicate]

I am interested in creating a neural network-based engine for chess. It uses a $8 \times 8 \times 73$ output space for each possible move as proposed in the Alpha Zero paper: Mastering Chess and Shogi ...
user avatar
  • 11
1 vote
1 answer
218 views

How to make the RL player a perfect/expert (tic-tac-toe/chess) player?

I asked a question related to tic-tac-toe playing in RL. From the answer, it seems to me a lot is dependent on the opponent (rightly so, if we write down the expectation equations). My questions are (...
user avatar
1 vote
2 answers
2k views

Minimax combined with machine learning to determine if a path should be explored

I have an idea for a new type of AI for two-player games with alternating turns, like chess, checkers, connect four, and so on. A little background: Traditionally engines for such games have used the ...
user avatar
1 vote
3 answers
1k views

Combining deep reinforcement learning with alpha-beta pruning

I will explain my question in relation to chess, but it should be relevant for other games as well: In short terms: Is it possible to combine the techniques used by AlphaZero with those used by, say, ...
user avatar
2 votes
3 answers
677 views

What is the definition of a heuristic function in the BayesChess paper?

I am reading BayesChess: A computer chess program based on Bayesian networks (Fernandez, Salmeron; 2008) It is a chess-playing engine using Bayesian networks. The following is mentioned about the ...
user avatar
  • 187
1 vote
1 answer
798 views

Does quiescence search even improve the minimax algorithm?

Consider this game state: d5 captures c6 Quiescence search returns about 8.0 as evaluation because after dxc6 and bxc6 Qxd6 would be played (then Qxd6 by black). A normal player would not play this ...
user avatar
  • 239
3 votes
1 answer
1k views

How is AlphaZero different from Stockfish or Rybka?

I don't know much about AI or chess engines, but what is the fundamental difference between AlphaZero and Stockfish or Rybka?
user avatar
2 votes
1 answer
728 views

Chess policy network [closed]

I am interested in making a simple chess engine using neural networks. I already have a fairly good value network but I can't figure out how to train a policy network. I know that Leela chess zero ...
user avatar
1 vote
1 answer
203 views

Is there a way of representing the minimax algorithm mathematically?

I have successfully figured out how the minimax algorithm works for a game like chess, where a game tree is used, and you assign a value to the terminal nodes and propagate that value up the tree. Is ...
user avatar
  • 457
3 votes
4 answers
2k views

If two perfect chess AI's played each other, would it always be a stalemate or would white win for an inherent first-move advantage?

In the circumstances of two perfect AI's playing each other, will white have an inherent advantage? Or can black always play for a stalemate by countering every white strategy?
user avatar
  • 131
6 votes
2 answers
3k views

What is the definition of rationality?

I'm having a little trouble with the definition of rationality, which goes something like: An agent is rational if it maximizes its performance measure given its current knowledge. I've read that ...
user avatar
1 vote
1 answer
394 views

Why was Go a harder game for an AI to master than Chess?

AI became superior to the best human players in chess around 20 years ago (when the 2nd Deep Blue match concluded). However, it took until 2016 for an AI to beat the Go world chess champion, and this ...
user avatar
4 votes
1 answer
2k views

How do you encode a chess move in a neural network?

In a neural network for chess (or checkers), the output is a piece or square on the board and an end position. How would one encode this? As far as I can see choosing a starting square is 8x8=64 ...
user avatar
  • 2,054
11 votes
3 answers
1k views

Why were Chess experts surprised by the AlphaZero's victory against Stockfish?

It was recently brought to my attention that Chess experts took the outcome of this now famous match as something of an upset. See: Chess’s New Best Player Is A Fearless, Swashbuckling Algorithm ...
user avatar
  • 6,067
1 vote
1 answer
559 views

What was the average decision speed pf Alpha Zero in the recent Stockfish match?

The match got a lot of press, and I doubt anyone is surprised that Alpha Zero crushed Stockfish. See: AlphaZero Destroys Stockfish in 100 Game Match To me, what's really salient is that "much like ...
user avatar
  • 6,067
4 votes
1 answer
683 views

How would you encode your input vector/matrix from a sequence of moves in game like tasks to train an AI? e.g. Chess AI?

I've seen data sets for classification / regressions tasks in domains such as credit default detection, object identification in an image, stock price prediction etc. All of these data sets could ...
user avatar
  • 63
4 votes
1 answer
700 views

Is there any value given to each chess piece in AlphaZero? [closed]

Recently, DeepMind's AlphaZero chess algorithm did better than the prior best chess software Stockfish. I read the paper Mastering Chess and Shogi by Self-Play with a General Reinforcement Learning ...
user avatar
  • 151
4 votes
1 answer
247 views

Why is chess still a benchmark for Artificial Intelligence?

Even though modern chess playing programs have demonstrated themselves to be as strong (or stronger) than even the best human players for nearly 20 years now (1997 when IBM's Deep Blue defeated the ...
user avatar
  • 143
0 votes
5 answers
988 views

Can an AI learn how to play chess without instructions?

Can an AI learn to play chess if you give it nothing but "the goal is to win" as starting criteria? If not, what is the minimum information the AI would need to be "seeded" with in order to learn to ...
user avatar
2 votes
2 answers
113 views

Trading off "Memory" vs "Optimization"

I've been researching the following topic. Or rather, I would like to but I can't find anything because I'm not sure what to look for. I am interested weather there are some concepts or models that ...
user avatar
  • 175
6 votes
4 answers
1k views

Is the play of strong Chess AI easily distinguishable from human play?

I don't play nearly enough Chess to be able to answer. For context, AlphaGo is stronger than the current strongest human player, but AlphaGo's game play has been cast as "inhuman" in the sense that ...
user avatar
  • 6,067
11 votes
3 answers
2k views

Why spend so much time and money to build AIs to play games?

I was reading about John McCarthy and his orthodox vision of Artificial Intelligence. To me, it seems like he was not very much in favour of resources (like time and money) being used to make AIs play ...
user avatar
-1 votes
3 answers
419 views

Are machine learning and self-learning really possible? [closed]

Some AI's, such as some chess players, are extremely well coded and have defeated humans in several matches. But I think that they won simply because computers can make calculations way faster than ...
user avatar
  • 25
2 votes
1 answer
143 views

Are strong Chess AI's local on mobile devices? [closed]

Just wondering about the architecture of strong Chess AI in a mobile, because networking is generally assumed by mobile developers, but not guaranteed.
user avatar
  • 6,067
5 votes
1 answer
582 views

Can games be solved without an evaluation function?

Fundamentally, a game-playing AI must solve the problem of choosing the best action from a set of possible actions. Most existing game AI's, such as AlphaGo, do this by using an evaluation function, ...
user avatar
  • 151
2 votes
1 answer
672 views

How could Deep Blue possibly cheat?

On Wikipedia we can read: Kasparov accused IBM of cheating and demanded a rematch. IBM refused and retired Deep Blue. What was the accusation and how was Deep Blue allegedly able to cheat?
user avatar
  • 10k
11 votes
5 answers
1k views

Are methods of exhaustive search considered to be AI?

Some programs do exhaustive searches for a solution while others do heuristic searches for a similar answer. For example, in chess, the search for the best next move tends to be more exhaustive in ...
user avatar
  • 2,313