Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Superdeterminism: The path we didn’t take

backreaction.blogspot.com

Post reply on HN
Print view View original

161–170 of 209 posts

Re: Superdeterminism: The path we didn’t take

#161

Post by vokep »

repsilat wrote: post #34
gliese1337 wrote: post #12

I find it strange that they claim that QM lacks any explanation for measurement. There are explanations for measurement. Not everyone agrees with any single one of them, but they definitely exist. Superdeterminism is one, but it's not the only one. It is really only the Copenhagen Interpretation of QM that has a measurement problem. Personally, I am on board with the "measurement is just entanglement" solution (it se…

I'm also a Many-Worldser, but I don't quite know how to answer the question "How does the Born Rule work?" That is, if "worlds" branch off with some amplitudes, and there's a "me" on this branch and another on that branch, how does it make sense to talk about the probability of "me me" being this one or that one? Empirically we know how to calculate it from the amplitude, but why is that more reasonable than any othe…

There is 100% probability that you are both. Why are you this one and not the other? Because you are not the other one. The other one wonders why it isn't you. The tautology should begin to be obvious. The answer is the same answer to why you are not me and I am not you. (The problem is the same across space within a timeline as it is across timelines)

Re: Superdeterminism: The path we didn’t take

#162

Post by vokep »

slashdev wrote: post #140
arnoooooo wrote: post #108

Earlier quoted context omitted.

The simplest interpretation is that there is no such thing as matter. Physicists are studying consciousness. It is actually the only explanation that is consistent with experience, since there is no experience outside of consciousness.

This is a new agey woo kind of idea that comes probably from a layman's interpretation of the effect of the observer on changing the quantum system being measured. However, it does not scale to macro systems. As Sam Harris put it when debating this idea with Deepak Chopra, the moon does not disappear when we're not looking at it. The universe was before consciousness and will continue to be afterwards.

But the moon does disappear. The moon as I understand it is a glowing disk with particular patterns on it. That disc and patterns only ever existed in my mind. When not looking at the moon, it goes back to being a very large number of particles near each other and nothing more.

Re: Superdeterminism: The path we didn’t take

#163

Post by gliese1337 »

edna314 wrote: post #156
gliese1337 wrote: post #149

Earlier quoted context omitted.

The math is the description of how it works. If you are not willing to accept any mathematical argument as a valid answer, you better explain what kind of answer you would accept as an alternative, 'cause it's not at all obvious that any other explanation could even exist.

I can argue in the same way in favor of Copenhagen. The math of Copenhagen is clearly defined, as the collapse of a wave function is the projection on eigenfunctions of the measuring device.

No, it isn't. The mathematics that describe the results of measurement are clearly defined, but what constitutes a measurement--i.e., when you need to apply that process--is not. There is no mathematics to back it up--it's purely ad-hoc.

That is what "measurement is just entanglement" provides.

Re: Superdeterminism: The path we didn’t take

#164

Post by edna314 »

gliese1337 wrote: post #163
edna314 wrote: post #156

Earlier quoted context omitted.

I can argue in the same way in favor of Copenhagen. The math of Copenhagen is clearly defined, as the collapse of a wave function is the projection on eigenfunctions of the measuring device.

No, it isn't. The mathematics that describe the results of measurement are clearly defined, but what constitutes a measurement --i.e., when you need to apply that process--is not . There is no mathematics to back it up--it's purely ad-hoc. That is what "measurement is just entanglement" provides.

I’m sorry, but I disagree. A measurement process is mathematically speaking an operator which acts on the wave function in Copenhagen interpretation and the collapse of the wave function is a projection onto the eigenfunctions of this operator. What would you be missing here? Instead of “measurement is just entanglement” I would claim “measurement is just an operator”. Since entanglement and operators can only be defined mathematically, I don’t see the point choosing one over the other.

Besides that, there is one point I don’t get about many worlds and entanglement theories: assuming that “measurement is just entanglement”, there are many possibilities how the observer gets entangled with the wave function it observes. How is it decided which entanglement (or which world) is the one we actually observe?

Re: Superdeterminism: The path we didn’t take

#165

Post by repsilat »

vokep wrote: post #161
repsilat wrote: post #34

Earlier quoted context omitted.

I'm also a Many-Worldser, but I don't quite know how to answer the question "How does the Born Rule work?" That is, if "worlds" branch off with some amplitudes, and there's a "me" on this branch and another on that branch, how does it make sense to talk about the probability of "me me" being this one or that one? Empirically we know how to calculate it from the amplitude, but why is that more reasonable than any othe…

There is 100% probability that you are both. Why are you this one and not the other? Because you are not the other one. The other one wonders why it isn't you. The tautology should begin to be obvious. The answer is the same answer to why you are not me and I am not you. (The problem is the same across space within a timeline as it is across timelines)

This sounds like the most reasonable conclusion until you actually start experimenting.

An example taken from another thread: Fire a laser of polarised light at a polarised filter angled at an offset of 30°. Measure the light on the other side of the filter and you'll see it's 75% as bright.

Now, you could say "We have worlds splitting independently for each photon that goes through or doesn't, and we're on all of those worlds," but that isn't explanatory -- it doesn't tell us why it looks like three quarters of the photons went through, as we always measure.

Re: Superdeterminism: The path we didn’t take

#166

Post by lmm »

GoblinSlayer wrote: post #135
lmm wrote: post #121

Earlier quoted context omitted.

> But, if you have probability then you have information, yes? What do you mean by that? Suppose you and I are far apart in space and are each going to toss a coin, but we somehow magically know that one of us will get heads and the other will get tails. Can we use this pair of magic coins to communicate information? Actually no.

The suggestion is to abuse information whether you get tails with probability 100% or 50%.

What do you mean?

Re: Superdeterminism: The path we didn’t take

#167

Post by AgentME »

trevyn wrote: post #70
kanzenryu2 wrote: post #60

Earlier quoted context omitted.

You may find this interesting https://www.lesswrong.com/posts/9cgBF6BQ2TRB3Hy4E/and-the-wi...

Wait, is MWI compatible with superdeterminism? That is, the sequence above seems to show that MWI is an effective way to calculate, but does not actually rule out superdeterminism underneath it — just like ordinary quantum mechanics is an effective way to calculate in certain domains of applicability but cannot effectively claim to be the “underlying reality”.

Superdeterminism was only proposed as an explanation for quantum "spooky action at a distance" without sacrificing the principle of locality. MWI already explains this weirdness without sacrificing the principle of locality, so there's no reason left to suggest superdeterminism if you accept MWI. MWI is compatible with and really is just Schrodinger's equation. Superdeterminism is compatible with practically anything, because it's just "what if the physics we know, plus something about the initial state happens to explain any spooky action at a distance [but MWI has no spooky action at a distance anyway]".

Re: Superdeterminism: The path we didn’t take

#168

Post by AgentME »

mikk14 wrote: post #133
GoblinSlayer wrote: post #94

Earlier quoted context omitted.

All theories are compatible with undetectable additions, but that's kinda whole point: MWI doesn't need them and works fine without them. But I believe superdeterminism has single world, so I don't see how it can emulate MWI.

I just had a funny thought. I don't know if it makes any sense, but it's funny to think about it. This is all made with my tongue firmly in my cheek, I don't know nearly enough about the topic to have any sort of confidence on what follows, so hold your downvotes :-) Let's assume MWI is correct. This means that, as far as I understand, the Schrodinger equation imposes that for each quantum event there is a forking in…

I believe your understanding of MWI is close enough, but I don't think your use of "superdeterminism" is correct. Superdeterminism is an attempt to give a local explanation for seemingly-non-local quantum phenomena. MWI is a specific local explanation for quantum phenomena, so there's no reason to add in a superdeterministic explanation for non-local phenomena, because there are no non-local phenomena in MWI that need any explaining away. I think you might be slightly mixing up superdeterminism with determinism (which is compatible with MWI).

Re: Superdeterminism: The path we didn’t take

#169

Post by lmm »

yyyk wrote: post #147
lmm wrote: post #122

Earlier quoted context omitted.

> it has no good explanation for why the Born rule applies at all. Indeed it doesn't, but the Born rule is much smaller and simpler than the notion of collapse that other approaches rely on. > A much more detailed way of saying about the same thing as the above sentence: https://backreaction.blogspot.com/2019/09/the-trouble-with-m... . This post doesn't seem to understand many-worlds. There is no concept of a privile…

>Indeed it doesn't, but the Born rule is much smaller and simpler than the notion of collapse that other approaches rely on. First, not all of the other approaches have a collapse. e.g. Bohm and a few lesser known others don't have it. Second, it's not clear Born is 'smaller and simpler' than the collapse of some other interpretations. e.g. in QBism collapse arises naturally from the theory and it would be weird if c…

> e.g. Bohm and a few lesser known others don't have it.

Bohm is just MWI with a bunch of extra epicycles.

> Second, it's not clear Born is 'smaller and simpler' than the collapse of some other interpretations. e.g. in QBism collapse arises naturally from the theory and it would be weird if collapse didn't happen.

I'd be interested in hearing more, but from a quick look I'd argue that the notion of an agent who should evaluate subjective probabilities is more complex than objective probability numbers.

> Her point is that what MWI tries to derive everything from the Schrödinger equation alone, but it still needs to solve its equivalent of the measurement problem (why can we only get certain results on the evolution of this branch?)

That's not true though. The branching behaviour is inherent in the Schrödinger equation.

Re: Superdeterminism: The path we didn’t take

#170

Post by lmm »

edna314 wrote: post #164
gliese1337 wrote: post #163

Earlier quoted context omitted.

No, it isn't. The mathematics that describe the results of measurement are clearly defined, but what constitutes a measurement --i.e., when you need to apply that process--is not . There is no mathematics to back it up--it's purely ad-hoc. That is what "measurement is just entanglement" provides.

I’m sorry, but I disagree. A measurement process is mathematically speaking an operator which acts on the wave function in Copenhagen interpretation and the collapse of the wave function is a projection onto the eigenfunctions of this operator. What would you be missing here? Instead of “measurement is just entanglement” I would claim “measurement is just an operator”. Since entanglement and operators can only be def…

> A measurement process is mathematically speaking an operator which acts on the wave function in Copenhagen interpretation and the collapse of the wave function is a projection onto the eigenfunctions of this operator. What would you be missing here?

When and how does that projection (which is non-unitary and otherwise ill-behaved) happen? What causes it to occur or not occur? How do you explain the results of the quantum eraser experiment, where the exact same physical process retroactively turns out to constitute a measurement or not depending on what happens in the future?

> Besides that, there is one point I don’t get about many worlds and entanglement theories: assuming that “measurement is just entanglement”, there are many possibilities how the observer gets entangled with the wave function it observes. How is it decided which entanglement (or which world) is the one we actually observe?

No, there is only one possible way to get entangled with what you observe. If you observe a spin-half particle that's in state (1/sqrt(2))(|up> + |down>), then afterwards the you-photon system is in state (1/sqrt(2))(|up>|you after observing up> + |down>|you after observing down>), exactly as if you were a normal physical object obeying the normal laws of physics.

The question you can and should still ask is "what should we expect the subjective experience of being in this state to be like". And to a certain extent, why the Born rule applies is still something to be explained (though I'm not aware of any alternative possibilities that would be well-behaved). But that's a much smaller and simpler question.

Post reply on HN