Um… I’m not sure I understand the question. I think you may have a different understanding of what Kegan’s stages are about than I do, and I’m not able to translate the question for that reason.
There can be no absolute proofs about anything in the macroscopic world. However, there are stronger and weaker arguments. And there are strong reasons to think there can be no optimal learning theory…
-
-
Philosophers of science spent decades trying to find a provably correct theory of induction. Instead, they found more and more reasons to think that no such theory is possible.
-
Half a century ago, the field switched to trying to understand more clearly why no such theory is possible, and (more importantly) why science often works anyway.
- 8 more replies
New conversation -
-
-
So let us make it narrower: given only a finite set of observations, each a finitely resolved bit vector, do you think someone is going to be able to prove that there can be no algorithm that has a higher likelihood to correctly predict the next observation than any other?
-
As I said: there are no proofs outside of mathematics. If these are genuinely observations, we are outside the scope of mathematics. That said: if the “observations” are produced by an adversarial hypercomputational oracle, then probably yeah you can prove no algorithm wins.
- 3 more replies
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.