Do electrons think? (1949) • Hacker News Board

Re: Do electrons think? (1949)

#31

Post by westoncb » Wed, Aug 15, 2018, 11:06 PM UTC

lmm wrote: post #28

westoncb wrote: post #26

Earlier quoted context omitted.

> Treating the wavefunction as physically real is perfectly reasonable. What's your justification for that? I remember Arthur Eddington pointing out that it could be interpreted as modeling our limited knowledge of a quantum system just as well as its 'physical reality'. But there's obviously justification for the first one, while the second one appears to be a large unjustified leap.

> What's your justification for that? Occam's razor. The wavefunction is the simplest construct that explains our experimental results (it appears in every interpretation to some degree or another, certainly in pilot wave theory where it is the pilot wave ), so it's simplest to assume it just is what's really happening. > I remember Arthur Eddington pointing out that it could be interpreted as modeling our limited kn…

> The wavefunction is the simplest construct that explains our experimental results (it appears in every interpretation), so it's simplest to assume it just is what's really happening.

You are addressing a straw man. I never said anything about an alternative to wavefunctions—I brought up how they should be interpreted. Additionally, your claim, "so it's simplest to assume it just is what's really happening" has no more justification to it than the first time you said it.

> Hidden variable approaches are incompatible with our experimental results ...

He wasn't talking about a hidden variable theory, just about how to interpret the wavefunction, holding all theoretical elements constant: does it model our knowledge of physical states, or the physical states themselves.

The first one is simpler because we already know we are observing the physical states through an intermediary and that our knowledge is limited; the second one introduces a new capability to the universe.

Re: Do electrons think? (1949)

#32

Post by stephengillie » Thu, Aug 16, 2018, 12:34 AM UTC

edem wrote: post #29

From what I understand "seeing" is an active proccess, since it needs light. If we want to observe small things we need more precise light, which means more energy. The problem is not that small particles are somehow become aware when we observe, but the observation itself adds energy to the system, thus alters it. Is this correct?

"Observing" means entangling with photons which have previously been entangled with the observed object. Thus, viewing an object means changing its state in such a way that you're temporarily also part of its state, forming an entropy gradient to gain energy/information from it.

Re: Do electrons think? (1949)

#33

Post by codethief » Thu, Aug 16, 2018, 12:35 AM UTC

tezthenerd wrote: post #27

codethief wrote: post #12

Earlier quoted context omitted.

> even though violations of Bell’s inequality have conclusively shown that the stage of our universe is nonlocal They have not. What they have shown is that the universe cannot be local and deterministic at the same time.

Bell's theorem rules out locality even for non-deterministic theories as well. Don't get determinism and realism confused...

I might not be using the same definitions as you are or maybe I'm getting something mixed up, so I'd be glad if you could elaborate on what exactly you mean by "determinism", "realism" and especially "locality".

(In my book, the Copenhagen interpretation is a non-realistic(∆), non-deterministic and local theory, which would contradict your statement.)

(∆) Assuming, of course, that the wave function is not an object of reality, as I think is standard.

Re: Do electrons think? (1949)

#34

Post by stephengillie » Thu, Aug 16, 2018, 12:39 AM UTC

Koshkin wrote: post #23

[Meta] A light, hard to read shade of gray. (Never understood why would anyone want to do that to the reader.)

HN uses the same poor text contrast to highlight unpopular opinions.

Re: Do electrons think? (1949)

#35

Post by tpfour » Thu, Aug 16, 2018, 12:53 AM UTC

crazygringo wrote: post #2

Note that, contrary to popular understanding, it isn't even established that quantum mechanics is indeterminate or not subject to strict cause-and-effect. It's indeterminate according to the mainstream Copenhagen interpretation, but de Broglie–Bohm theory [1] ("pilot wave theory") is an interpretation that is entirely deterministic, and its assumptions result in exactly the same final equations as the Copenhagen one.…

I think you are abusing language here, probably not intentionally, and it distracts from the actual issue. You seem to understand the difference but for those who don't, here's a crucial distinction that's often overlooked by laypeople:

The Schrödinger (Dirac) equation itself is deterministic. Given a wavefunction at time t=t0, you can calculate its evolution to time t=t1. What is indeterminate is the observation of the observable. The distribution of observations is entirely determined. Only the result of a single observation is indeterminate: in the limit of infinite measurements, you would get the probability distribution back which itself is determinate. The various "interpretations" concern themselves with what happens at the time of measurement. For example, the Copenhagen interpretation says that the wavefunction, which was evolving deterministically, mutates into a new wavefunction which itself evolves deterministically. This act of going from one to the other is considered indeterminate and is given the name of "collapse". However the dynamics of each wavefunction is entirely determinate.

Re: Do electrons think? (1949)

#36

Post by jmcmichael » Thu, Aug 16, 2018, 12:57 AM UTC

thraway180306 wrote: post #20

jmcmichael wrote: post #15

Earlier quoted context omitted.

Fine, here's another from Ronald Hensen at Delft: Experimental loophole-free violation of a Bell inequality using entangled electron spins separated by 1.3 km 2015 https://arxiv.org/abs/1508.05949v1 "...This result rules out large classes of local realist theories, and paves the way for implementing device-independent quantum-secure communication and randomness certification. "

This is something different. Zeilinger paints himself disproving realism alone and has a well known philosophical stance/bias here. Local realism testing here however is about being both local and realist, and pilot waves aren't local to begin with.

Yeah, that makes sense. Do you have an example of Zeilinger's writing where he discusses his philosophical stance?

Is a philosophical stance something you believe that a physicist shouldn't have?

Re: Do electrons think? (1949)

#37

Post by tbabb » Thu, Aug 16, 2018, 1:18 AM UTC

crazygringo wrote: post #2

Note that, contrary to popular understanding, it isn't even established that quantum mechanics is indeterminate or not subject to strict cause-and-effect. It's indeterminate according to the mainstream Copenhagen interpretation, but de Broglie–Bohm theory [1] ("pilot wave theory") is an interpretation that is entirely deterministic, and its assumptions result in exactly the same final equations as the Copenhagen one.…

De Broglie-Bohm is probably not the correct theory, since (if I am not mistaken) it's not reconcilable with relativity, and has a host of other problems which led its original proponents to abandon it.

Copenhagen is also probably not correct, though it is still frustratingly mainstream. Copenhagen has the well known problem of lacking a physical definition of "observer", but it also has conflicts with delayed choice quantum eraser experiments (which suggest that the wavefunction never collapses), and contradictions that arise when you try to locate the proposed wave collapse in spacetime.

The Everett many-worlds interpretation is gaining favor, and has none of the problems of the other two pictures. In the Everett view, scientists and experiments and instruments and universes are ensembles of particles like any other, and enter superpositions upon interacting with other systems in superposition. In a sense, every possible outcome of a quantum measurement actually happens in due proportion to the probabilities described by quantum mechanics, but upon becoming entangled with the experiment result, the different outcomes can no longer influence or measure each other. It is the most ontologically minimal theory, since it proposes no other events or laws beyond the already-known laws of quantum mechanics.

Re: Do electrons think? (1949)

#38

Post by effie » Thu, Aug 16, 2018, 1:46 AM UTC

panic wrote: post #11

I wonder what Schrödinger would have thought of the Free Will Theorem: https://arxiv.org/pdf/quant-ph/0604079.pdf

I think he would not like it. The theorem is underwhelming and uses misleading attention-seeking terminology. A more appropriate way to state it is, I think, this: if experiment settings violate determinism, its results have to violate determinism as well. To the author's credit, they say as much in the introduction themselves - "I saw you put the fish in".

Re: Do electrons think? (1949)

#39

Post by IAmGraydon » Thu, Aug 16, 2018, 1:48 AM UTC

stephengillie wrote: post #32

edem wrote: post #29

From what I understand "seeing" is an active proccess, since it needs light. If we want to observe small things we need more precise light, which means more energy. The problem is not that small particles are somehow become aware when we observe, but the observation itself adds energy to the system, thus alters it. Is this correct?

"Observing" means entangling with photons which have previously been entangled with the observed object. Thus, viewing an object means changing its state in such a way that you're temporarily also part of its state, forming an entropy gradient to gain energy/information from it.

That is completely incorrect.

Re: Do electrons think? (1949)

#40

Post by effie » Thu, Aug 16, 2018, 1:53 AM UTC

tbabb wrote: post #37

crazygringo wrote: post #2

Note that, contrary to popular understanding, it isn't even established that quantum mechanics is indeterminate or not subject to strict cause-and-effect. It's indeterminate according to the mainstream Copenhagen interpretation, but de Broglie–Bohm theory [1] ("pilot wave theory") is an interpretation that is entirely deterministic, and its assumptions result in exactly the same final equations as the Copenhagen one.…

De Broglie-Bohm is probably not the correct theory, since (if I am not mistaken) it's not reconcilable with relativity, and has a host of other problems which led its original proponents to abandon it. Copenhagen is also probably not correct, though it is still frustratingly mainstream. Copenhagen has the well known problem of lacking a physical definition of "observer", but it also has conflicts with delayed choice…

In the Everettian interpretation, experiments do not have results (!) and it is not clear what probability even means, much less how the Born rule and observed statistics arise. That's because it is not a theory about our world, but about an imagined world where the only things existing is the psi function.

> every possible outcome of a quantum measurement actually happens in due proportion to the probabilities described by quantum mechanics

This cannot work, because those probabilities are defined by a basis in Hilbert space and that basis is defined by details of the experiment, which does not actually happen in the Everettian view.