Different things, no? Kauffman emphasizes that evolution of the genome is non-ergodic, which is certainly true. Friston only needs, presumably, the evolution of brain states to be ergodic. That's plausible, it's a much smaller space.
-
-
-
My understanding here is worse than sketchy, however. (And I don't know why/if the free energy principle would really rely on ergodicity. It's not a closed system, there should be attractors.)
- 15 more replies
New conversation -
-
-
If the science turns out to show beyond reasonable doubt that living systems are ergodic, to what extent would this compromise other aspects of the enactivist program? In other words, to what extent are you theoretically committed to this bet going your way?
-
Good question. I don't think enactive ideas stand or fall with this issue. Part of the problem here is exactly what "ergodic" means. It may be that "nonstationary" is more relevant. (Ergodic may be compatible with nonstationary, but I need physicists to say)
- 2 more replies
New conversation -
-
-
Isn't ergodicity the basis of fractal geometry ? If living systems are based on ergodicity they should possess fractal like features.
-
Interesting. I don't know enough about this. But wouldn't it be enough for living systems to be a mix of ergodic and non-ergodic processes for them to have fractal features?
- 1 more reply
New conversation -
-
-
Isn’t ergodicity a far too precise concept to be either applicable or inapplicable to life? I wonder what does it even mean in a non-statistical physics fashion? (If it’s directionally applicable, does anyone have a working definition for this context?)
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Friston uses ergodic ensemble as a working definition of a single organism from within his formal framework Kauffman is looking at evolution and penetration of the adjacent possible In general, biology is nonergodic, but the approximation might be reasonable in certain places
-
Would you say that the working approximation of even visiting across a range of states, requires dynamic exchange with and movement in a place? i.e. if a system is deprived of active inference options, too few states get engaged, and the surprisal approach becomes less optimal.
- 6 more replies
New conversation -
-
-
i need to spend more time w Friston's work, but it seems to me that if the internal states have been able to map the regularity of the base flow (represented by the external states) to the external states, then the internal+external states will be in an ergodic attracting set. 1/
-
if the internal states haven't accomplished that yet, they have some learning to do. it doesn't seem to me that a ton hangs on actually reaching optimality, but i'm only one paper in. also not sure how to scale up from subsystems. fascinating stuff tho. 2/pic.twitter.com/Q7zL8hBgxm
- 1 more reply
New conversation -
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.