It’s always fun to read papers about board games, especially after ironing out the kinks that make them unenjoyable.

I would love to see such a paper about Puerto Rico (my all time favorite board game), #christmaswishlist

> ironing out the kinks that make them unenjoyable.

I can't help look for clues to real problems, if the paradigm fits. The role of camouflage got my attention in this paper, as for cybersecurity recon, if the known cards are open capabilities, but the mystery cards are vulnerabilities then the analysis formalises something defensive hackers do know... that honeynets with lots of plausible cover traffic really slow attackers down.

Now apply these to the Nash equilibrium in the bar scene of A Beautiful Mind. Imagine an AI that could perform such analysis in real time in a social context. Such a thing could give a great advantage to players of imperfect information games like "love" and "politics".

An AI that could simply detect existing games that are being played, where they are so often invisible, would be a revolution by itself.

The most promising attempt to apply AI to real-world imperfect information games that I am aware of, is the AI startup Strategy Robot: https://www.strategyrobot.ai/

Of course, they're focusing on military applications. It seems like a very cool company and I'm interested to see how they do.

mods: typo in title ("srategies")