Questions tagged [game-theory]
For questions regarding the use of the mathematical theory of games (Von Neumann, Morgenstern, Nash etc) in AI. For questions about the use of AI in game design and gaming, use [gaming].
59
questions
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 ...
- 155
9
votes
4
answers
4k
views
Is there any board game where a human can still beat an AI?
Significant AI vs human board game matches include:
chess: Deep Blue vs Kasparov in 1996,
go: DeepMind AlphaGo vs Lee Sedol in 2016,
which demonstrated that AI challenged and defeated professional ...
- 10.5k
6
votes
3
answers
1k
views
Perfect play in information incomplete games
As titled, is there such thing as perfect play (or at least "perfectly optimal") in a game with incomplete information? Or at least a proof as to show why there cannot?
Naively (and seemingly ...
- 2,111
6
votes
2
answers
791
views
Are perfect and imperfect information games modelled as fully and partially observable environments, respectively?
In perfect information games, the agent can see all the moves performed in the past. Besides, it can observe the next action that will be put into practice by the opponent.
In this case, can we say ...
- 161
6
votes
1
answer
182
views
Interesting examples of discrete stochastic games
SGs are a generalization of MDPs to multiple agents. Like this previous question on MDPs, are there any interesting examples of zero-sum, discrete SGs—preferably with small state and action spaces? I'...
- 347
5
votes
1
answer
968
views
What's the optimal policy in the rock-paper-scissors game?
A deterministic policy in the rock-paper-scissors game can be easily exploited by the opponent - by doing just the right sequence of moves to defeat the agent. More often than not, I've heard that a ...
- 1,161
5
votes
1
answer
325
views
Why do zero-sum perfect information games satisfy the conditions of Von Neumann's theorem?
The Von Neumann's Minimax theorem gives the conditions that make the max-min inequality an equality.
I understand the max-min inequality, basically ...
- 167
4
votes
3
answers
6k
views
How to calculate the optimal placements for settlements in Catan without an ML algorithm?
Is it possible to calculate the best possible placements for settlements in Catan without using an ML algorithm?
While it is trivial to simply add up the numbers surrounding the settlement (highest ...
- 41
4
votes
1
answer
2k
views
Why is tic-tac-toe considered a non-deterministic environment?
I have been reading about deterministic and stochastic environments, when I came up with an article that states that tic-tac-toe is a non-deterministic environment.
But why is that?
An action will ...
- 193
4
votes
2
answers
448
views
What is the difference between genetic algorithms and evolutionary game theory algorithms?
What is the difference between genetic algorithms and evolutionary game theory algorithms?
- 1,743
4
votes
1
answer
587
views
How can an AI play Flow Free?
The game "Flow Free" in which you connect coloured dots with lines is very popular. A human can learn techniques to play it.
I was wondering how an AI might approach it. There are certain rules of ...
- 2,216
4
votes
2
answers
515
views
How to define or represent evil in logic
Is there any well defined method to define or represent evil in abstract logic, binary or AI form?
Video games method of representing evil is relative to the player context (thus subjective, and not ...
- 169
4
votes
1
answer
424
views
What is the state-space complexity of Spades?
AI reached super-human level in many complex games, including imperfect information games such as six-player no-limit Texas hold’em poker. However, it still did not reached that level in Trick-taking ...
- 413
4
votes
0
answers
376
views
What is the state of the art AI training technique for imperfect information 2 player turn based games?
As far as I can tell (correct me if I'm wrong), Alphazero (with MCTS and neural network heuristic function RL) is the state of the art training method for turn based, deterministic, perfect ...
- 171
4
votes
1
answer
335
views
Can minimax be used when both players want to increase their score?
If both the players want to increase their score (by selecting the highest or best cost path), can this be done using the minimax algorithm, or are there other algorithms for this purpose?
- 41
3
votes
4
answers
5k
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?
- 131
3
votes
2
answers
5k
views
How to make minimax optimal?
By optimal I mean that:
If max has a winning strategy then minimax will return the strategy for max with the fewest number of moves to win.
If min has a winning strategy then minimax will return the ...
- 64
3
votes
1
answer
446
views
Why does our AI play worse at even levels of depth?
We are building an AI to play a board game. Leaving aside the implementation, we noticed that it plays worse when we set an even (2,4,6,...) level of depth. We use a minimax depth-first strategy.
Do ...
- 33
3
votes
1
answer
254
views
Is there a way to predict points on a map?
I have a data set with historical information of some events (let's say event A and event B),these events describe the discovery of land mines, the coordinates of the event and the date of the event; ...
- 73
3
votes
0
answers
151
views
What are some strong algorithms for Perfect Information, Deterministic Multiplayer Games?
I have a series of games with the following properties:
3 or more players, but purely non-cooperative (i.e., no coalition forming);
sequential moves;
perfect information;
deterministic state ...
- 131
3
votes
0
answers
79
views
Negative counterfactual regret
I am reading the paper Regret Minimization in Games with Incomplete
Information on CFR algorithm.
On page 4, the paper defines $R^{T,+}_{i,\text{imm}}=\max\{R^{T}_{i,\text{imm}}, 0\}$ after equation (...
- 131
3
votes
0
answers
79
views
Is there a measure of AI relative strength, modified by resources?
For instance, Strength/Size$\times$Speed, where size and speed refer to memory and processing.
We now have very strong, narrow AI, but they tend to run on fast hardware without volume restrictions.
To ...
- 6,235
2
votes
2
answers
389
views
What is the difference between game theory and machine learning?
What is the difference between game theory and machine learning?
I had gone through the papers Deep Learning for Predicting
Human Strategic Behavior, by Jason Hartford et al., and When Machine ...
- 1,743
2
votes
2
answers
203
views
AI with conflicting objectives?
A recent question on AI and acting recalled me to the idea that in drama, there are not only conflicting motives between agents (characters), but a character may themselves have objectives that are in ...
- 6,235
2
votes
1
answer
221
views
Is there any flaw to this solution to the One shot prisoner's dilemma
I wrote a solution to the one shot Prisoner's dilemma:
Introduction
My solution applies to a prisoner dilemma involving two people (I have neither sufficient knowledge of the prisoner'...
- 165
2
votes
0
answers
112
views
When to model decision-making problem as single agent vs multi-agent problem?
I understand the goals and purposes of RL in the case of a single agent and the underlying model, i.e. MDPs, for RL problems (or sequential decision making with uncertainty in general).
My question is ...
- 121
2
votes
1
answer
225
views
How are the Bellman optimality equations and minimax related?
Is the philosophy between Bellman equations and minimax the same?
Both the algorithms look at the full horizon and take into account potential gains (Bellman) and potential losses (minimax).
...
- 171
2
votes
0
answers
37
views
Impact or applications of introducing attention in deep networks modelling multi-agent systems
I have been reading quite a lot about the research progress in the domain of self attention-based neural networks that were introduced by Google Inc. in their paper titled "Attention is all you ...
- 21
2
votes
0
answers
59
views
Understanding the nature of psychological defense system by artificial intelligence
Do you think psychological defense system, for example, repression, regression, reaction, formation, isolation, undoing, projection, introjection, sublimation, etc., could be created by artificial ...
- 151
2
votes
0
answers
121
views
What is the difference between evolutionary game theory and meta-heuristics?
Here is a list of meta-heuristic algorithms
Ant colony optimization,
Ant lion optimizer,
Artificial bee colony algorithm,
Bat algorithm,
Cat swarm optimization,
Crow search algorithm,
Cuckoo ...
- 1,743
2
votes
0
answers
102
views
How does Friend-or-Foe Q-learning intuitively work?
I read about Q-Learning and was reading about multi-agent environments. I tried to read the paper Friend-or-Foe Q-learning, but could not understand anything, except for a very vague idea.
What does ...
- 21
2
votes
0
answers
125
views
Strong and Weak Dominance Table
I have this table, 2 agents and I want to find for each agent if any action is strongly or weakly dominated. This is the table:
Now, i've found a solution but I'm not sure if it's correct. So for let'...
- 21
1
vote
1
answer
142
views
Is neural fictitious self play violating off-policy theorem
I was reading the NFSP player from D. Silver, and I'm somewhat confused by the algorithm:
In particular, given that we sample an action according to best response ($\sigma = \epsilon-\text{greedy}(Q)$...
- 2,473
1
vote
2
answers
278
views
Can Alpha–Beta be used on symmetric zero sum games?
This question was asked in an AI exam. How would you answer such question?
- 111
1
vote
1
answer
239
views
Shouldn't the utility function of two-player zero-sum games be in the range $[-1, 1]$?
In Appendix B of MuZero, they say
In two-player zero-sum games the value functions are assumed to be bounded within the $[0, 1]$ interval.
I'm confused about the boundary: Shouldn't the value/...
- 471
1
vote
3
answers
171
views
What approaches are there to apply AI to global economic processes?
Today, AI is mainly driven by own-profit-oriented companies (e.g. Facebook, Amazon, Google). Admittedly, there's a lot of AI in the health sector (even in the public health sector) and there's a lot ...
1
vote
2
answers
700
views
Why do iterative deepening search start from the root each iteration in the context of the minmax-algorithm?
Consider the graph below for an understanding on how IDS work.
Now my question is:
why do IDS start at the root every iteration, why not start at the previously searched depth in the context of minmax?...
- 11
1
vote
2
answers
703
views
Would AlphaZero work just with a value network?
There is a nice post about the intuition why AlphaZero works.
One of the advantages of using a policy network in the games where a perfect simulator is available (such as chess) is to save computation ...
- 181
1
vote
1
answer
1k
views
How exactly is Monte Carlo counterfactual regret minimization with external sampling implemented?
I have read many papers, such as this or this, explaining how external sampling works, but I still don't understand how the algorithm works.
I understand you divide $Q$, which is the set of all ...
- 11
1
vote
1
answer
731
views
Optimal mixed strategy in two player zero sum games
I am currently studying game theory based on Peter Norvig's 3rd edition introduction to artificial intelligence book. In chapter 17.5, the two player zero sum game can be solved by using the $\textbf{...
- 1,291
1
vote
1
answer
306
views
Compressing text using AI by sending only prediction rank of next word
Is there any effort made to compress text (and maybe other media) using prediction of next word and thus sending only the order number of the word/token which will be predicted on the client side
i.e
...
- 111
1
vote
1
answer
82
views
Can programs like AlphaGo be said to be means of dealing with computational intractability?
I mean this in the sense that Go is unsolvable but AlphaGo seems able to make choices that are consistently more optimal than a human player's choices.
It is my understanding that Game Theory ...
- 6,235
1
vote
2
answers
117
views
Is there serious game-theoretic work on AI risk and alignment?
My background is in political economy and game theory. I am interested in the discussion on AI risk and alignment, but I have so far failed to find work on this that seriously engages with classic ...
1
vote
0
answers
88
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 ...
1
vote
0
answers
71
views
Book/course recommendation on game theory application to multi-agent system (reinforcement learning)
Is there any great game theory book or course that discusses the application of game theory to modern reinforcement learning or multi-agent systems? Or a classic reference book that can help me get a ...
- 471
1
vote
0
answers
44
views
AI for games which involve social intelligence. Games like warewolf where players must persuade, charm, threaten etc
I'm looking for any introductory/accessible reading on AI that can play games which involve social intelligence.
Games like poker, where you might bait someone into overcommiting their hand or ...
- 111
1
vote
0
answers
533
views
What is the size of 6-players no limit Texas holdem Poker?
What is the number of game states/information sets in 6-players, no limit, Texas Holdem?
A year ago, Pluribus reached a super-human level in 6-players no limit Holdem Poker. I am interested in the ...
- 413
1
vote
1
answer
483
views
Find the expected reward in an expectimax-based dice rolling game?
I have this question that I'm kinda stuck on.
It's a game scenario in which we set up an expectimax tree. In the game, you have 3 dice with sides 1-4 that you roll at the beginning. Then, depending on ...
- 21
1
vote
0
answers
103
views
Minimax algorithm with only partial visibility
I'm trying to implement the minimax algorithm with alpha beta pruning on a game that works like this:
Player 1 plays (x1, y1).
Player 2 can only see the x-value (x1) that Player 1 played (and not ...
1
vote
0
answers
277
views
Is Monte Carlo tree search guaranteed to converge to the optimal solution in two player zero-sum stochastic games?
I'm aware that convergence proofs for Monte Carlo tree search exist in the case of deterministic zero sum games and Markov decision processes.
I have come across research which applies MCTS to zero-...
- 11