local optimums are dangerous - they can get civilizations collapsed. Use evolutionary algorithms & ensure enough level of mutation is present in the society ;)https://twitter.com/Plinz/status/941522164586557440 …
-
-
Replying to @hackermill
For societies, the real bitch is how to find Nash equilibria that are compatible with things not going to shit. Offsetting incentives requires governance, incentives of government need to be aligned with creating global optima, you need adaptive error correction etc.
2 replies 1 retweet 3 likes -
Replying to @Plinz
once Nash equilibria is found then what is the point for a society to exist beyond this point?
1 reply 0 retweets 0 likes -
-
Replying to @Plinz
heh, so the only way to ensure human survival as a race is to never reach Nash equilibria and just spend eternity searching and deliberately turning away once (if) one is seen on the horizon :)
1 reply 0 retweets 0 likes -
Replying to @hackermill
Misunderstanding! The Nash equilibrium is reached when all individual agents have optimized their behavior according to their incentives (whatever they think these are). Often this local equilibrium is not producing the globally best outcome, which is why you need governance.
2 replies 0 retweets 0 likes -
Replying to @Plinz
I mean even a local maximum can produce better results than Nash equilibrium, right? Probably they can be even the same (local maximum and equilibrium)pic.twitter.com/DC2KKUcrLH
1 reply 0 retweets 0 likes
Any local optimum of the global system corresponds to a state where all individual agents have reached Nash equilibria. (That is simplified, of course; actual economies will almost never become stationary.)
-
-
Replying to @Plinz
btw, does that mean that agents once reached Nash equilibria have to be stationary?
0 replies 0 retweets 0 likesThanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.