Hacker News new | ask | show | jobs
by joycian 2750 days ago
Can you supply a non-deterministic process, where the non-determinism is fundamental and not a way to quantify a lack of information about the system?
3 comments

Yep. That's the question. Some quantum processes are indeterministic. However, one might ponder if it is because of ignorance of the inner-workings on the said process.

One also might conclude that non-determinism is ignorance or lack of information. It is an unanswered question as far as I know.

Anything that is best described by quantum mechanics.
https://en.m.wikipedia.org/wiki/Superdeterminism

What do you think of ideas like this then?

I think it's a boring theory that doesn't pass Ockham's razor. For the universe to be superdeterministic you'd have to encode its whole evolution somewhere. That is more complicated that the normal interpretation of QM but doesn't explain more.
> I think it's a boring theory that doesn't pass Ockham's razor. For the universe to be superdeterministic you'd have to encode its whole evolution somewhere

The Cellular Automaton Interpretation of Quantum Mechanics, by 't Hooft https://arxiv.org/abs/1405.1548

That's really interesting, I like CAs a lot. However the author himself states in the introduction that Bell's inequality and related theorems very strongly suggest that his theory is not correct.
> However the author himself states in the introduction that Bell's inequality and related theorems very strongly suggest that his theory is not correct.

It's widely believed that Bell's theorem and other no-go theorems preclude a local, hidden variable interpretation of QM. The author merely mentions that he can reconcile existing results with this local, hidden variable interpretation via "superdeterminism".

Isn't it simpler? There are just a few rules and everything works? For the normal QM you have to get some randomness from somewhere else.

edit: I mean I agree it's problematic for experimental science, but I think Occam's razor cuts the other way in this instance.

edit2: Isn't the entire history encoded anyway in the current state of the universe in the deterministic case. I don't understand what the problem is there.

To get rid of the randomness you have to have prepared the result of every decision that is currently attributed to randomness in a big lookup table.
Why would that be necessary if there are local deterministic propagation rules, that have been there since the beginning?
That's only according to some interpretations of quantum mechanics. (I.e. Copenhagen) Other interpretations, such as Many Worlds or Pilot Wave are deterministic.
Many worlds is deterministic, but not in the same sense that people might think of, right. The evolution of the wavefunction (which never collapses like in Copenhagen) is deterministic, but in any given world, random draws are appearing.
The big bang/creation