2
$\begingroup$

Game Theory is a sub-branch of economics. In fact, Economics Stack Exchange has many questions tagged with . We also have some questions tagged with this tag. However, to me, given that I'm not an expert in this topic (I only know about minimax and similar adversarial search algorithms applied to solve games like tic-tac-toe in the context of AI), it's not clear to what extent questions about game theory should be on-topic on Artificial Intelligence SE.

So, I was wondering whether we should explicitly mention "Game Theory" under our topics on our on-topic page. If not, do you think that something related to game theory that is also related to AI (such as "adversarial search") should be explicitly mentioned in our on-topic page? Note that our on-topic page already mentions "search".

$\endgroup$
5
$\begingroup$

I'd say that questions related to game-theory can very often end up being on topic on AI.SE, but it probably doesn't need to be explicitly listed as being on-topic in its entirety. Game theory does show up in various areas of AI (not just like minimax for search in games, but also for like any other kinds of multi-agent interactions, and probably some other areas I don't know enough about), so anything related to that should be on-topic in my opinion. I wouldn't explicitly list game theory as a whole as a topic, because if someone really has a complex, pure game theory question outside of any other AI context, they'd probably be better served on an economics or math website.

$\endgroup$
1
  • $\begingroup$ I skimmed some pretty interesting recent category theoretic game theory papers a couple of years ago, and I feel like that approach can subsume pretty much everything! ("Game" and "player" seem to become arbitrary, b/c it's all just functions.) I agree with you on the complex questions being mor suitable for math or economics—that's a pretty specialize area, with different concern than our concerns. $\endgroup$
    – DukeZhou Mod
    Aug 31 at 3:27
0
$\begingroup$

I personally think all of the game theories can comment at a much more fundamental level than search, because I see it as the root of all rational decision making, which is the basis for intelligence (utility in an action space.)

I don't know that we need an explicit on-topic reference, but, in view of genetic algorithms/evolutionary game theory, I think we might want to consider it...

(Because I'm doing a database program, I'm currently thinking about game theoretic approaches to data storage, structure, and management. I don't know if it will be fruitful, but it will be interesting!)

I'd want to specify that we're interested in game theory in regard to things like search, genetic algorithms, and rational agents, not specific real world economic questions like those explored by think tanks.

$\endgroup$
1
  • $\begingroup$ (I also think there is probably a huge amount of bias in the applied game theory space, where you can twist logic to support preferred, non-cooperative strategies. You see this most clearly in attempted refutations of Nash, with simple logic that doesn't hold in the most basic iterated dilemmas, because there is a faction that rejects cooperative behavior.) $\endgroup$
    – DukeZhou Mod
    Aug 31 at 3:33

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .