Qbit Information Transfer achieved by TU Delft

[Reaver225]

Teleporting information achieved by TU Delft
A key step toward a "quantum internet"
June 2, 2014


Illustration representing quantum information teleportation between diamonds (credit: TU Delft)

Teleporting people through space, as in Star Trek, is impossible by the laws of physics, but researchers at TU Delft‘s Kavli Institute of Nanoscience have succeeded in teleporting information.

Using quantum entanglement, they transferred the information contained in a quantum bit in a diamond to a quantum bit in another diamond three meters away, without the information having traveled through the intervening space.

The results were published in Science May 29.

Hanson’s research group produces qubits using electrons in diamonds. “We use diamonds because ‘mini prisons’ for electrons are formed in this material whenever a nitrogen atom ["nitrogen vacancy center"] is located in the position of one of the carbon atoms. The fact that we’re able to view these miniature prisons individually makes it possible for us to study and verify an individual electron and even a single atomic nucleus.

“We’re able to set the spin (rotational direction) of these particles in a predetermined state, verify this spin, and subsequently read out the data. We do all this in a material that can be used to make chips out of. This is important as many believe that only chip-based systems can be scaled up to a practical technology.”

Quantum internet

The researchers say this development is an important step towards a network for communication between future ultra-fast quantum computers — a “quantum internet” that will enable secure information transfer.

“The distance in our tests was three meters, but in theory the particles could be on either side of the universe,” said the head of the research project, Prof. Ronald Hanson. “The unique thing about our method is that the teleportation is guaranteed to work 100%. The information will always reach its destination, so to speak. And, moreover, the method also has the potential of being 100% accurate.”

Hanson is planning to repeat the experiment this summer over a distance of 1300 meters, with diamonds located in various buildings on TU Delft’s campus. This experiment could be the first that meets the criteria of the “loophole-free Bell test,” and could provide the ultimate evidence to disprove Einstein’s rejection of entanglement. Various other research groups are currently striving to be the first to realize a loophole-free Bell test, which is considered to be the Holy Grail in quantum mechanics.

Abstract of Science paper

Realizing robust quantum information transfer between long-lived qubit registers is a key challenge for quantum information science and technology. Here, we demonstrate unconditional teleportation of arbitrary quantum states between diamond spin qubits separated by 3 m. We prepare the teleporter through photon-mediated heralded entanglement between two distant electron spins and subsequently encode the source qubit in a single nuclear spin. By realizing a fully deterministic Bell-state measurement combined with real-time feed-forward quantum teleportation is achieved upon each attempt with an average state fidelity exceeding the classical limit. These results establish diamond spin qubits as a prime candidate for the realization of quantum networks for quantum communication and network-based quantum computing.

Source: http://www.kurzweilai.net/teleporting-i ... y-tu-delft

Abstract: http://www.sciencemag.org/content/early ... ce.1253512

This is big news, isn't it? Exactly how they did it is also something to be wondered about, if they actually pulled it off.

[Channel72]

If something like a "quantum Internet" is actually possible via exploiting quantum entanglement to send information, then the ramifications are truly insane. It would mean we could have instant communication across interstellar distances.

[Darmalus]

Even if the information transfer is still speed of light, having "perfect" wireless communication links would improve the internet in ways I can barely imagine.

[Ziggy Stardust]

I admit this is not a subject I know terribly much about, so could anyone in the know explain statements like this:

“The distance in our tests was three meters, but in theory the particles could be on either side of the universe,” said the head of the research project,

I am just ... wary of such extreme, inherently unverifiable statements, especially in pop sci sources about physics and the like. I mean, you can spout just about whatever bullshit you want as long as you append "in theory" to it. It just bothers me when people making grandiose claims without backing them up specifically.

[Channel72]

As crazy as that claim is - they don't really need to back it up, assuming they're actually capable of experimentally reproducing what they say they've accomplished. Quantum entanglement is inherently a phenomenon which acts at arbitrary distances, instantaneously. 3 meters or 3 light-years, it doesn't matter. Yes, it violates relativity, yes it's fucked up, yes, Einstein didn't like it. But subsequent experiments have empirically verified that it does, for whatever reason, actually happen - somehow, the state of one particle can be instantaneously affected by measuring the state of another particle, even if the two particles are on opposite sides of the galaxy (or even Universe.) How the actual information is transmitted is unknown.

Whether or not this can be exploited to somehow instantaneously send information is still very debatable, and the general consensus is that it can't... but these guys claim they did it. Whether they're full of shit remains to be seen, I guess.

[Iroscato]

To quote my ever-eloquent father when I showed him the article..."Well fuck me in my stink-hole".
This, if true, has so many consequences and applications it makes my head spin. Instantaneous interplanetary communication, perfect signal coverage, and of course teleportation.
And if it's truly instantaneous, how about interplanetary travel? We would need a ship to go to Mars to establish a colony and whatever machine necessary for the teleportation to occur, but we would make ships practically obsolete.
If it's true, and that's a big if, this could be or be the first step to being the biggest breakthrough of the 21st century

[Purple]

Aren't you jumping the gun just a little bit? Sure we could send a glorified radio to Mars or Venus or the other edge of the galaxy and chat with someone over there in real time. But it's still a glorified radio. And it will do nothing to actually help us put someone there to chat with, let alone keep him sustained.

[Iroscato]

Aren't you jumping the gun just a little bit? Sure we could send a glorified radio to Mars or Venus or the other edge of the galaxy and chat with someone over there in real time. But it's still a glorified radio. And it will do nothing to actually help us put someone there to chat with, let alone keep him sustained.

Absolutely I'm jumping the gun, though I prefer the term 'Excitedly thinking ahead way ahead'. But the teleportation side of things absolutely would be. They would allow us to transmit resources from any facility in the solar system without waiting for a ship that needs time and fuel to move.
I appreciate this is way off in the future, if it's even possible, but when it comes to discoveries like this I turn into an excited little puppy.

[Kuroneko]

If something like a "quantum Internet" is actually possible via exploiting quantum entanglement to send information, then the ramifications are truly insane. It would mean we could have instant communication across interstellar distances.

It doesn't mean any such thing.

Quantum entanglement is inherently a phenomenon which acts at arbitrary distances, instantaneously. 3 meters or 3 light-years, it doesn't matter. Yes, it violates relativity, yes it's fucked up, yes, Einstein didn't like it.

Quantum entanglement doesn't violate relativity. It does not imply any sort of interaction. Two particles are entangled if and only if some of their measurement outcomes are not independent.

Suppose I take a pair of shoes and put them in different boxes and give them to Alice and Bob, respectively. No matter how far apart they are, if Alice looks in her box, she'll instantly know what type of shoe Bob would find should he choose to look in his, and vice versa. The measurement outcomes are not independent. The only thing quantum mechanics introduces is that states are allowed to be superpositions.

But subsequent experiments have empirically verified that it does, for whatever reason, actually happen - somehow, the state of one particle can be instantaneously affected by measuring the state of another particle, even if the two particles are on opposite sides of the galaxy (or even Universe.) How the actual information is transmitted is unknown.

No, it's not affected at all. If Alice and Bob each have a particle entangled with the other, then whatever is done to one of them, nothing whatsoever happens to the other, i.e. there is no experiment whatsoever that will tell you what if anything happened to the other.

Whether or not this can be exploited to somehow instantaneously send information is still very debatable, and the general consensus is that it can't... but these guys claim they did it. Whether they're full of shit remains to be seen, I guess.

If you actually read the article, you'll find they don't claim anything remotely like what you ascribe them to.

Suppose Alice wants to send Bob some information. To do it securely, they arrange to have a random key, known to both. Alice then uses the key to scramble the information she wants to send, sends the scrambled info, and Bob can use the shared key to unscramble it. In this classical version, Bob will then have a copy of the original information, while Alice still has her copy. In the quantum version, the entangled particles serve as the key and the process destroys Alice's information. Hence it's "teleportation" in the sense that Alice had the information before and Bob has it after, i.e. there are never multiple copies.

N.B. There is no mention of anything about "instantaneous". Actually, since quantum teleportation protocol requires sending the 'scrambled' info with a classical communication channel, it's no faster in terms of time delay than already existing communication. The advantages is that it sends qubits instead of classical bits and that it can do so securely (QT is analogous to a one-time-pad in classical crypto, but better).

[Channel72]

Thanks for the explanation Kuroneko.

Can you elaborate a bit on why the "popular interpretation", at least, of quantum entanglement seems to contradict what you're saying? Most laymen call to mind Einstein's "spooky action at a distance" when thinking of quantum entanglement, and the popular literature on it claims that it does, in fact, act at arbitrarily long distances. For example, an excerpt form the wiki article:

The seeming paradox here is that a measurement made on either of the particles apparently collapses the state of the entire entangled system – and does so instantaneously, before any information about the measurement could have reached the other particle (assuming that information cannot travel faster than light). In the quantum formalism, the result of a spin measurement on one of the particles is a collapse into a state in which each particle has a definite spin (either up or down) along the axis of measurement. The outcome is taken to be random, with each possibility having a probability of 50%. However, if both spins are measured along the same axis, they are found to be anti-correlated. This means that the random outcome of the measurement made on one particle seems to have been transmitted to the other, so that it can make the "right choice" when it is measured. The distance and timing of the measurements can be chosen so as to make the interval between the two measurements spacelike, i.e. from any of the two measuring events to the other a message would have to travel faster than light. Then, according to the principles of special relativity, it is not in fact possible for any information to travel between two such measuring events – it is not even possible to say which of the measurements came first, as this would depend on the inertial system of the observer. Therefore the correlation between the two measurements cannot appropriately be explained as one measurement determining the other: different observers would disagree about the role of cause and effect.

The way I read this, is that measuring one particle directly affects the spin of another, even if the other particle is (theoretically) at the other end of the Universe.

I understand your argument about the pair of shoes. If there's only two possible, mutually exclusive, outcomes, then if you get outcome A, you know the other particle got outcome B - so in this sense, no real "information" was transmitted. But still, how do we avoid concluding that somehow a physical phenomena has propagated at faster than light speeds? The outcome of measuring particle B must always be the opposite of the outcome of measuring particle A, therefore particle B must somehow "know" the state of particle A.

[StandingInFire]

No, it's not affected at all. If Alice and Bob each have a particle entangled with the other, then whatever is done to one of them, nothing whatsoever happens to the other, i.e. there is no experiment whatsoever that will tell you what if anything happened to the other.

I found the explanations and "Alice/Bob" metaphors are really bad at explaining anything to me on quantum entanglement or qbits. Since you seem to know stuff about it could you clarify some stuff for me.

From what I gather you are saying:
If Particle A is observed in State 1 = Particle B is observed it will also be in State 1.
If Particle A is manipulated to be in State 2 = Particle B is observed in a state independent of Particle A. i.e. manipulation breaks entanglement (permanently or temporarily?).

As for qbits they can be in 3 states A, B, and a superposition of A & B, but from what I read the superposition of A & B is a result of the Schrodinger effect so only exists till you observe it, in which case I don't understand how that state would affect anything as it would not exist when it is actually being read but might have an effect if one is tying to manipulate it. Is that correct?

[Ziggy Stardust]

As crazy as that claim is - they don't really need to back it up, assuming they're actually capable of experimentally reproducing what they say they've accomplished. Quantum entanglement is inherently a phenomenon which acts at arbitrary distances, instantaneously. 3 meters or 3 light-years, it doesn't matter. Yes, it violates relativity, yes it's fucked up, yes, Einstein didn't like it. But subsequent experiments have empirically verified that it does, for whatever reason, actually happen - somehow, the state of one particle can be instantaneously affected by measuring the state of another particle, even if the two particles are on opposite sides of the galaxy (or even Universe.) How the actual information is transmitted is unknown.

So you basically just ignored my post to repeat the exact thing I was criticizing?

WHY do they not need to back it up? That's ridiculous. This is science. You can't make statements without some way to back them up. What about the ability to transmit this information over 3 meters implies that 3 light-years or any other arbitrary distance is possible? Yes, it may be possible in theory, but why can they just say that distance is completely meaningless when they have not actually done any experiments to verify that this is the case? This is exactly the problem I was complaining about. There is no verifiability or accountability in such a theory, which makes it empirically pointless. What work has been done that implies quantum entanglement completely ignores distance, as implied? Just spouting nonsense like "oh, well, in theory it does" is just that - nonsense. That isn't how fucking science works.

This is like me pricking my finger with a pin, noting I feel the sensation instantly, then saying in theory nerves allow for instantaneous transfer of information over any distance.

[Kuroneko]

Can you elaborate a bit on why the "popular interpretation", at least, of quantum entanglement seems to contradict what you're saying? Most laymen call to mind Einstein's "spooky action at a distance" when thinking of quantum entanglement, and the popular literature on it claims that it does, in fact, act at arbitrarily long distances.

Einstein is a big name, so people repeat his sayings even when he's wrong. Experimental violation of Bell's inequality proves that Einstein dream of 'hidden variables' determining outcomes that QM described probabilistically proved him wrong.

The way I read this, is that measuring one particle directly affects the spin of another, even if the other particle is (theoretically) at the other end of the Universe.

affect, v: produce an effect or change in. Saying that measurement of particle A affects the spin of particle B means that the following two situations give (at least statistically) different results:
-- Measuring spin of B if A has been measured.
-- Measuring spin of B if B has not been measured.
... but they don't. At all. If you have B, there is absolutely nothing you can do it that gives even the slightest indication of whether or not A has been measured. In other words, measurement of particle A produces no change in particle B.

I understand your argument about the pair of shoes. If there's only two possible, mutually exclusive, outcomes, then if you get outcome A, you know the other particle got outcome B - so in this sense, no real "information" was transmitted. But still, how do we avoid concluding that somehow a physical phenomena has propagated at faster than light speeds?

If it makes absolutely no measurable difference, then saying that a physical phenomenon has propagated is a very strange thing to conclude. Perhaps what's tripping you up is that you believe that "the state" of the particle is some universally shared objective fact common to all observers. Rather, it's just a description of an observer's information about the system.

Say Alice has her shoebox. She doesn't know what's in it, but she knows a random shoe from a pair has been put in it with equal probability, and that Bob has the other shoe in his own shoebox. They both reckon: if Bob looks in his shoebox, he'll find a left shoe with 50% chance and a right shoe with 50% chance. They repeat this experiment a gazillion times, and find it statistically agrees with this.

Now they do it again, only Alice looks in her shoebox, but doesn't tell Bob. Suppose she finds a [left OR right] shoe. Then she reckons: if Bob looks in his shoebox, he'll find a [right OR left, resp.] shoe with 100% probability. Bob's probability has collapsed! But Bob doesn't know any of this, so he reckons same as before: 50% chance left, 50% chance right. They repeat this experiment a gazillion times. Bob has a 50/50 split as before, but Alice is perturbed by the fact that Bob's outcome is fully predicted by the information she finds in her shoebox.

Alice: I could predict your result by looking, so some physical phenomenon has propagated from my shoebox to yours!
Bob: Well, we repeated both experiments a gazillion times, and I saw no difference at all between them, so it must not be much of a physical phenomenon.

Meanwhile, the reality is actually quite simple: in the second experiment, Alice's description of Bob's shoebox is different from Bob's description of Bob's shoebox because she knows more.

---

I found the explanations and "Alice/Bob" metaphors are really bad at explaining anything to me on quantum entanglement or qbits.

I'm sorry to hear that, but the point is only to illustrate ordinary, classical correlation.
Quantum entanglement is just that plus allowing things to be in superposition.

From what I gather you are saying:
If Particle A is observed in State 1 = Particle B is observed it will also be in State 1.
If Particle A is manipulated to be in State 2 = Particle B is observed in a state independent of Particle A. i.e. manipulation breaks entanglement (permanently or temporarily?).

That's essentially correct.

As for qbits they can be in 3 states A, B, and a superposition of A & B, but from what I read the superposition of A & B is a result of the Schrodinger effect so only exists till you observe it, in which case I don't understand how that state would affect anything as it would not exist when it is actually being read but might have an effect if one is tying to manipulate it. Is that correct?

No, it's not correct. For example, take an electron in equal superposition of spin-up and spin-down along, say, the x-axis:
*** |ψ〉 = (1/√2)( |x,up〉 + |x,dn〉)
For this state, if one measures the spin along the x-axis, one will get spin-up with 50% chance and spin-down with 50% chance ((1/√2)² = 50%). However the same state also has:
*** |ψ〉 = |z,up〉
Measuring spin along the z-axis gives spin-up with 100% chance. In other words, the same state is both a superposition and a definite state, depending on which observables we're talking about. Of course, not every superposition is a definite state of another observable, but this counter-example is enough to show that superpositions somehow being "not real" is a mistake.

[StandingInFire]

No, it's not correct. For example, take an electron in equal superposition of spin-up and spin-down along, say, the x-axis:
*** |ψ〉 = (1/√2)( |x,up〉 + |x,dn〉)
For this state, if one measures the spin along the x-axis, one will get spin-up with 50% chance and spin-down with 50% chance ((1/√2)² = 50%). However the same state also has:
*** |ψ〉 = |z,up〉
Measuring spin along the z-axis gives spin-up with 100% chance. In other words, the same state is both a superposition and a definite state, depending on which observables we're talking about. Of course, not every superposition is a definite state of another observable, but this counter-example is enough to show that superpositions somehow being "not real" is a mistake.

Is there a predefined method for assigning the axis to the particle?
If not then:
Since the particle is in 3D the state is dependent on essentially the angle you are using to look at the particle.
If there is:
The particle can spin freely about its axis which can induce such a state as the spin "rotates".

[Kuroneko]

I'm not sure I understand your question.

Pick an arbitrary direction in space, which we'll call the z-axis. If you measure the electron spin along this axis with some measurement device**, sometimes you'll get spin-up and sometimes you get spin-down. If you select just the spin-up ones and throw away the spin-down ones, you will have electrons in |z,up〉 state. Now you can take those and do stuff with them, e.g.,
(1) If you measure those electrons again along the same axis (with a second identical device), all of them will register spin-up again. The electrons are in a definite state with respect to the z-axis.
(2) If instead you rotate measure along a different axis, you will find that sometimes they register spin-up and sometimes spin-down.
The point is that a definite spin state along one axis is actually a superposition along a different axis. But it's the same state.

**The ur-example of actually doing this is passing the electron through a Stern-Gerlach magnet, which has a strong inhomogneous magnetic field. The magnetic field deflects the electron toward opposite directions, conventionally called up and down. The fact that they go in different directions makes it easy to post-select the state you want and throw away the other.

[StandingInFire]

Pick an arbitrary direction in space, which we'll call the z-axis.

That is what I was asking, as in are the axis defined arbitrarily or was there some method for defining direction (for example if it was an electron orbiting a nucleus you could define the orbital path as x-axis, and y-axis pointing towards the nucleus, and that would also give you a z-axis and then the electron could rotate freely about its own internal axis within that defined space). Seems like its just arbitrary from what I can gather.

[Channel72]

I understand your argument about the pair of shoes. If there's only two possible, mutually exclusive, outcomes, then if you get outcome A, you know the other particle got outcome B - so in this sense, no real "information" was transmitted. But still, how do we avoid concluding that somehow a physical phenomena has propagated at faster than light speeds?

If it makes absolutely no measurable difference, then saying that a physical phenomenon has propagated is a very strange thing to conclude. Perhaps what's tripping you up is that you believe that "the state" of the particle is some universally shared objective fact common to all observers. Rather, it's just a description of an observer's information about the system.

Say Alice has her shoebox. She doesn't know what's in it, but she knows a random shoe from a pair has been put in it with equal probability, and that Bob has the other shoe in his own shoebox. They both reckon: if Bob looks in his shoebox, he'll find a left shoe with 50% chance and a right shoe with 50% chance. They repeat this experiment a gazillion times, and find it statistically agrees with this.

Now they do it again, only Alice looks in her shoebox, but doesn't tell Bob. Suppose she finds a [left OR right] shoe. Then she reckons: if Bob looks in his shoebox, he'll find a [right OR left, resp.] shoe with 100% probability. Bob's probability has collapsed! But Bob doesn't know any of this, so he reckons same as before: 50% chance left, 50% chance right. They repeat this experiment a gazillion times. Bob has a 50/50 split as before, but Alice is perturbed by the fact that Bob's outcome is fully predicted by the information she finds in her shoebox.

Alice: I could predict your result by looking, so some physical phenomenon has propagated from my shoebox to yours!
Bob: Well, we repeated both experiments a gazillion times, and I saw no difference at all between them, so it must not be much of a physical phenomenon.

Yes, I see what you're saying.

Still, I'm not certain if the shoe analogy really works. No analogy is perfect, of course, but if wave function collapse is analogous to Alice opening the box and looking at the shoe, the analogy breaks down in the sense that an actual left shoe or right shoe exists in the box regardless of whether Alice ever looks at it. But with wave function collapse, the spin value of the observed particle doesn't actually come into existence until the particle is observed and the wave collapses. This goes for the entangled peer particle as well - which may very well be off in Andromeda by now. So, the act of Alice observing the particle (and collapsing the wave function), causes Bob's particle to have an opposite spin - even though Bob's particle may be light years away.

Again, I understand your shoe analogy, but isn't it a bit faulty? In the case of the shoe, either a left shoe or right shoe is definitely physically inside Alice/Bob's box, before the box is ever opened (or even regardless if it's opened at all, i,e, opening the box does not cause a shoe with the property of "leftness" or "rightness" to come into existence). But with the particle, it's the act of observation itself that collapses the wave function and gives us a particle with a spin value.

[Channel72]

So you basically just ignored my post to repeat the exact thing I was criticizing?

WHY do they not need to back it up? That's ridiculous. This is science. You can't make statements without some way to back them up. What about the ability to transmit this information over 3 meters implies that 3 light-years or any other arbitrary distance is possible? Yes, it may be possible in theory, but why can they just say that distance is completely meaningless when they have not actually done any experiments to verify that this is the case? This is exactly the problem I was complaining about. There is no verifiability or accountability in such a theory, which makes it empirically pointless. What work has been done that implies quantum entanglement completely ignores distance, as implied? Just spouting nonsense like "oh, well, in theory it does" is just that - nonsense. That isn't how fucking science works.

This is like me pricking my finger with a pin, noting I feel the sensation instantly, then saying in theory nerves allow for instantaneous transfer of information over any distance.

The affect of observing one particle in an entangled pair has empirically been show to cause the peer particle to have an opposite spin "instantaneously" (or at least, at speeds faster than light.)

If I'm understanding him correctly, Kuroneko is basically saying that it doesn't really matter, because no "state" is actually being transferred, - it's just that one observer instantaneously knows something about the properties of a far away particle.

But, I don't see how Kuroneko's interpretation is compatible with quantum theory, which clearly states that a particle literally doesn't exist (and thus has no spin value) in any particular state, until is is observed and the wave function (superimposed particle) collapses. So observing particle A causes particle A to suddenly "spin up", which then instantaneously transfers state to particle B (even if particle B is light years away) so that now particle B has spin "down".

As far as I can see, Kuroneko is basically saying that both particle A and B already had opposite spin values before they were observed, and the only thing that has changed is that now the observer of particle A is informed. But that's not what quantum theory actually implies.

[Kuroneko]

People tend to mystify QM to such degree...

Still, I'm not certain if the shoe analogy really works. No analogy is perfect, of course, but if wave function collapse is analogous to Alice opening the box and looking at the shoe, the analogy breaks down in the sense that an actual left shoe or right shoe exists in the box regardless of whether Alice ever looks at it.

Right, except that wasn't part of the analogy. The more exact analogy is in the collapse of the classical probability distribution once Alice look in the box.

I've said above that the intrinsically quantum addition is the allowance of superposition; my point that that this is not conceptually relevant to entanglement per se, but rather QM generally: the intuitive weirdness of superposition applies just as well to single particle states as to entanglement. Entanglement itself is simply the translation of joint probability distributions into quantum language: two particles are entangled iff some of their measurement outcomes are not independent.

But with wave function collapse, the spin value of the observed particle doesn't actually come into existence until the particle is observed and the wave collapses.

In as much as this statement is even meaningful, it's true of probabilities generally, whether classical or quantum. Wavefunction 'collapse' is simply conditioning probability on an outcome. That's conceptually the same in both classical and quantum probabilities.

Actually, one can make this much stronger: classical mechanics has wavefunction formulation as well, living in a complex Hilbert space and measurement projecting ('collapsing') onto the eigenspace of the measurement result, all just like in QM. Notably, in the standard Hilbert space formalism, the definition of entanglement is completely blind as to whether we're dealing with classical or quantum wave functions. The difference between classical wavefunctions and quantum wavefunctions is the physical role of phase, which in CM causes attempts of making nontrivial superpositions to fail. But that has nothing to do with wavefunction collapse.

This goes for the entangled peer particle as well - which may very well be off in Andromeda by now. So, the act of Alice observing the particle (and collapsing the wave function), causes Bob's particle to have an opposite spin - even though Bob's particle may be light years away.

Everything you said about wavefunction collapse is just as true in classical mechanics as in quantum mechanics. It's a completely general feature of probabilities. So why in the world do you claim this conclusion is special for the QM case?

---

The affect of observing one particle in an entangled pair has empirically been show to cause the peer particle to have an opposite spin "instantaneously" (or at least, at speeds faster than light.)

What experiment are you thinking of? Can you provide one?

To be clear, my position is that the above is just plain false. Have Alice and Bob measure their entangled particles (in the singlet spin state, for simplicity), repeated numerous times.
v1) Alice doesn't measure her particle. Bob measures his particle. He gets 50% spin-up, 50% spin-down.
v2) Both measure their particles. They each get 50% spin-up, 50% spin-down, and individually look the same as (v1). However, comparing their results shows that they directly correspond to each other. This is true no regardless of who measures first.

Clearly, we have correlation. What empirical evidence do you have of causation?

The case against is very simple: the individual results are same regardless of who measures first, and neither has any indication whatsoever regarding whether or not the other performed the measurement. You shouldn't use words like "cause" or "physical phenomenon" or "act" for something that makes no measurable difference at all. Of course, when the results are compared, they are correlated, but as any statistician will tell you, correlation is not necessarily causation.

The actual physical cause for their correlation is that the particles were produced by the same process in the past; here, one that conserves spin angular momentum.

If I'm understanding him correctly, Kuroneko is basically saying that it doesn't really matter, because no "state" is actually being transferred, - it's just that one observer instantaneously knows something about the properties of a far away particle.

The state is a description of the observer's knowledge.

But, I don't see how Kuroneko's interpretation is compatible with quantum theory, which clearly states that a particle literally doesn't exist (and thus has no spin value) in any particular state, until is is observed and the wave function (superimposed particle) collapses.

For simplicity, and as an adequate example, let's talk about measuring some particle's position. Suppose you do this and find get some particular result. What standard quantum mechanics says is that the question of where the particle was before the measurement is physically meaningless. That only has a vague resemblance to what you said here. Quantum mechanics teaches us to just talk about measurement outcomes. Whether something "literally doesn't exist" before it's measured is not be a physical claim, but rather a philosophical one.

As to whether my position is compatible quantum theory, yes, I believe so: all throughout the above discussion I've repeatedly interpreted questions like whether "a physical phenomenon has propagated" when Alice measured her particle in terms of measurement outcomes. Does it make any detectable difference whatsoever to Bob? No: Bob has no indication through his particle as to what, if anything, Alice does, or doesn't, do. Therefore, no "physical phenomenon has propagated".

As far as I can see, Kuroneko is basically saying that both particle A and B already had opposite spin values before they were observed, and the only thing that has changed is that now the observer of particle A is informed. But that's not what quantum theory actually implies.

If you mean opposite definite spin values, then no, I've claimed no such thing. (It's certainly possible for it to be such in a some particular experiment, but it's not true in general.)
If "opposite spin values" allows superpositions, then yes, that's true. Their oppositeness is due to whatever process prepared them to be such.

[Channel72]

People tend to mystify QM to such degree...

Still, I'm not certain if the shoe analogy really works. No analogy is perfect, of course, but if wave function collapse is analogous to Alice opening the box and looking at the shoe, the analogy breaks down in the sense that an actual left shoe or right shoe exists in the box regardless of whether Alice ever looks at it.

Right, except that wasn't part of the analogy. The more exact analogy is in the collapse of the classical probability distribution once Alice look in the box.

I've said above that the intrinsically quantum addition is the allowance of superposition; my point that that this is not conceptually relevant to entanglement per se, but rather QM generally: the intuitive weirdness of superposition applies just as well to single particle states as to entanglement. Entanglement itself is simply the translation of joint probability distributions into quantum language: two particles are entangled iff some of their measurement outcomes are not independent.

Again, I understand your position - but I'm a bit baffled how you're comfortable completely ignoring the singularly defining feature of wave function collapse when comparing it to the collapse of a standard, classical probability distribution. The difference between a standard probability distribution collapse, and a quantum probability distribution (wave function) is that the mechanism behind a quantum collapse is observation. There is no correlation between observation and a classical probability distribution collapse. If I roll a die, and it lands on 5 - it actually landed on 5, regardless of whether I (or anyone else) checks the result. However, QM indicates that no result whatsoever is even defined until someone observes it.

If the act of observation is said to be the mechanism for the probability distribution collapse, then I still don't see how we're avoiding the "action at a distance" of quantum entanglement, since neither particle even has a spin to speak of, until somebody observes either particle in the pair. Therefore, if entangled particles A and B are 1,000 lightyears apart, and someone observes particle A, particle B suddenly has a defined spin.

I understand the pragmatic nature of your position - but the fundamental "mystifying" aspect of QM has always been the connection between observation and wave function collapse, which doesn't seem to impress you very much. I suppose whether or not quantum entanglement is "interesting" depends on your philosophical beliefs about the underlying ontological implications of QM... but that is really the age old question, I suppose. I imagine you prefer the Copenhagen interpretation.

The affect of observing one particle in an entangled pair has empirically been show to cause the peer particle to have an opposite spin "instantaneously" (or at least, at speeds faster than light.)

What experiment are you thinking of? Can you provide one?

I wasn't thinking of any particular experiment - I was under the impression that local realism had been experimentally disproved on multiple occasions. But a simple web search comes up with: http://www.nature.com/nature/journal/v4 ... E-20130509 . And here's a nice abstract (pdf).

[Kuroneko]

Again, I understand your position - but I'm a bit baffled how you're comfortable completely ignoring the singularly defining feature of wave function collapse when comparing it to the collapse of a standard, classical probability distribution.

I'm not ignoring it. Quite the opposite: I'm saying it's exactly the same. Because it's the same, I don't see any reason to mystify that particular feature of QM (now, superposition, on the other hand, is a genuine difference).

The difference between a standard probability distribution collapse, and a quantum probability distribution (wave function) is that the mechanism behind a quantum collapse is observation.

Again: this is a general feature of probabilities, whether classical or quantum. If you learn new information about the system, you update the probabilities accordingly.

There is no correlation between observation and a classical probability distribution collapse.

This is mistaken. Let's examine how what state collapse even means.

Glossing over most of the technicalities, a (pure) state is a vector |ψ〉 in some complex Hilbert space, and physical observables correspond to Hermitian operators on that Hilbert space. If you measure some observable, you get one of its eigenvalues, and the state is in the eigenspace of the eigenvalue you obtained. In other words, measurement randomly projects the state vector onto the eigenspace of the result. This is "state collapse". The wave function is just the representation of the state vector in some particular basis formed by eigenstates of some observable(s).

Here's the thing: everything in the preceding paragraph is applicable to both classical and quantum physics. Exactly the same in both, not just as some vague analogy. The essential difference between CM and QM is the physical role of complex phase (which is related to how states change in time: for QM in the Schrödinger picture, the states obey the Schrödinger equation, while CM states obey the Liouville equation instead), and some incidental differences about which physical observable corresponds to which mathematical operators. But those differences have nothing to do with state collapse whatsover.

If I roll a die, and it lands on 5 - it actually landed on 5, regardless of whether I (or anyone else) checks the result.

This is a somewhat vague philosophical statement. While I agree that classical mechanics allows some philosophical interpretations that are not compatible with quantum mechanics, I'm simply saying that neither wavefunction collapse nor entanglement are responsible for that.

Ok, take a simple classical point-particle, no spin/internal structure. Unlike the quantum version, it can have definite position and momentum simultaneously. If the particle state has definite position and momentum, they will stay definite, and the classical particle will trace out some particular trajectory in position-momentum space. Furthermore, if the particle state is a combination of different trajectories, then they don't interfere with one another.

The existence of starkly defined trajectories and their non-interference with one another is what allows your philosophical claims about classical mechanics: if the particle position/momentum is not definite, but rather some probability distribution, one can say that it "really is" something definite and "you just don't know it". Specifically, here's the list of featured of classical mechanics that enable this:
-- (1) Ability to have simultaneously definite position and momentum. (Reason: position and momentum operators commute in CM, while they don't commute in QM.)
-- (2) Ability to have starkly defined trajectories: states of definite position and momentum stay definite. (This is a statement about time evolution of states.)
-- (3) The non-interference of those different trajectories from one another.
The reason for the last one is that in CM, the wavefunction phase and modulus evolve independently from another another, which again has to do with the structure of the Liouville equation, as opposed to the Schrödinger equation.

I understand the pragmatic nature of your position - but the fundamental "mystifying" aspect of QM has always been the connection between observation and wave function collapse, which doesn't seem to impress you very much.

This is a sentiment often endorsed by popular descriptions of QM, but it's just plain wrong.

Look at the above list responsible for your philosophical interpretation of "it was something definite all along and you just didn't know it yet". The first is a disagreement about which physical observables correspond to which mathematical operators (esp. commuting vs non-commuting), while the latter two both stem from a different time evolution equation. "The connection between observation and wave function collapse" is simply not relevant. That's common to both QM and CM anyway.

I wasn't thinking of any particular experiment - I was under the impression that local realism had been experimentally disproved on multiple occasions.

Yes, that's true. However, I don't agree that this justifies your earlier claim.

[Kuroneko]

I should probably give some explicit references for the above.

Brief background: quantum mechanics is due to the work of a great number of people, but the person most responsible for its mathematical formalization in terms of Hilbert spaces is von Neumann, most relevantly here including the treatement of measurement as projective state collapse. He was also involved in doing much the same thing to classical mechanics in the early 1930s, which is known as Koopman-von Neumann mechanics.

For people well-versed in quantum mechanics, there's an interesting review of KvN classical mechanics on arXiv:
On Koopman-von Neumann Waves
On Koopman-von Neumann Waves II
The second part contains a discussion of superposition in classical mechanics: it is found that as far as every classical observable is concerned (using the definition transplanted from the much more commonly-known Hamiltonian formulation of CM), classical superpositions are indistinguishable from mixed states. That's essential to the "it was definitely this all along and we just didn't know it" philosophical interpretation of classical mechanics.

Tangent: even without an explicit KvN treatement, the fact that classical mechanics must have a Hilbert space formulation is justified by some of the comparatively (that is, compared to KvN) well-known techniques of quantization. Spoiler

One learns from Hamiltonian mechanics that obserables are functions over the phase space and that they form a commutative Lie algebra with the Poisson bracket as the Lie bracket. To quantize it, one can impose a deformation to make position and momentum non-commutative, which forms a Heisenberg algebra. The question of where Hilbert spaces and operators come into play is answered by the Gel'fand-Naimark theorem, which states that any sufficiently "nice" algebra is essentially the algebra of some suitable operators on a Hilbert space. Then the picture becomes obvious: if we can get a Hilbert space formulation after deformation, it must also be possible before deformation, because the algebra of classical observables before deformation is all that but even "nicer" (i.e., is also commutative).

Although since KvN formulation already already exists, those considerations not necessary.

I believe that this more than adequately supports my prior contentions:
-- Mystefying state collapse due to observation is a mistake. It's a general feature of probabilities, both in quantum mechanics and classical mechanics.
-- Mystefying entanglement is a mistake. It's a general feature of non-independent joint probabilities, again the same in both.
-- However, quantum superposition is genuinely different from anything in classical mechanics, and indeed quite strange to classically-trained intuition. If there's anything in QM worth mystefying, it is that.
-- Therefore, explaining quantum entanglement as "just ordinary correlations except that things are allowed to be in superposition" is perfectly valid.

[Channel72]

You're obviously more well-versed in quantum mechanics than I am, but even I know that your interpretation is certainly controversial.

The second part contains a discussion of superposition in classical mechanics: it is found that as far as every classical observable is concerned (using the definition transplanted from the much more commonly-known Hamiltonian formulation of CM), classical superpositions are indistinguishable from mixed states. That's essential to the "it was definitely this all along and we just didn't know it" philosophical interpretation of classical mechanics.

This, in particular, is very disturbing.

If you're correct, the Schrödinger's cat thought experiment doesn't even require a Geiger counter. We could just hook up some classical probability distribution apparatus to the hammer and poison flask, like a mechanical coin flipping device, or an ordinary computer that uses some internal source of entropy to produce a (pseudo) random number - and we'd have the same counter-intuitive problem of a dead/alive cat before it is observed.

But of course this is completely silly because nobody truly believes the collapse of a classical probability distribution is non-deterministic. The outcome of a coin flip, after all, depends on things like air pressure, gravitational pull, wind factor, the exact motion of the hand, etc. Even a Turing machine is theoretically incapable of producing a non-deterministic event, which is why truly random number generators rely on quantum phenomena.

[Kuroneko]

You're obviously more well-versed in quantum mechanics than I am, but even I know that your interpretation is certainly controversial.

I'm claiming that quantum states describe probabilities, which is about as mainstream as it gets. (It should be emphasized Einstein's "spooky action at a distance" is far away from mainstream. Pretty much everyone agrees that he was mistaken.)

This, in particular, is very disturbing.

Why do you find it disturbing? Out of physicists that think the quantum measurement problem as a genuine problem (so basically excluding Copenhagen, "shut up and calculate", and a few others, which together is very probably the plurarity and might be the majority), most think that quantum decoherence solves the measurement problem (this includes MWI and some others). What I said essentially means that classical superpositions always lack coherence, and so if there's a quantum measurement problem that's solved by decoherence, then in CM it is self-solving!

If you're correct, the Schrödinger's cat thought experiment doesn't even require a Geiger counter. We could just hook up some classical probability distribution apparatus to the hammer and poison flask, like a mechanical coin flipping device, or an ordinary computer that uses some internal source of entropy to produce a (pseudo) random number - and we'd have the same counter-intuitive problem of a dead/alive cat.

That doesn't follow. The issue with the Schrödinger's cat is accounting for the emergence of classicality from quantum mechanics, i.e. the loss of coherence of quantum superposition. I've said that above in CM, are always indistinguishable from mixed states as far as any classical observable is concerned. That means that in CM, "superposition of dead and alive" is really just the same as "some chance of being alive, some chance of being dead."

But of course this is completely silly because nobody truly believes the collapse of a classical probability distribution is non-deterministic. A coin flip, after all, depends on things like air pressure, gravitational pull, the exact motion of the hand, etc.

Suppose you measure a large classical system, such a cat in Schrödinger's torture-box, and find her alive. Reminder:

The state is a description of the observer's knowledge.

What "collapsed" is your knowledge of the cat. Before, it was uncertain, a probability distribution over "alive" and "dead" possibilities. Now, it is certain.

I've also repeatedly agreed above that in the classical case, you can interpret this situation as the cat was alive before you looked. Where my disagreement with you lies is regarding why one can always do that in CM, but not always in QM. According to you, this is because there's something special about state collapse during measurement, while my contention was this is because there's something special about quantum superpositions themselves. The kicker is that Schrödinger's cat-torture experiment was explicitly arguing that there's something quite "wrong" about quantum superpositions, as a "half-dead, half-alive cat" shouldn't be allowed. And you're telling me that I'm silly in claiming that the essential difference between CM and QM lies in their treatment of superpositions?

---

Finally, regarding what is and isn't controversial, I'll take this bit from wikipedia:

Michael Nielsen counters: "at a quantum computing conference at Cambridge in 1998, a many-worlder surveyed the audience of approximately 200 people... Many-worlds did just fine, garnering support on a level comparable to, but somewhat below, Copenhagen and decoherence."

MWI actually also uses quantum decoherence to solve the measurement problem, but it isn't the only one, so the majority of responders actually fall into two classes:
(1) The measurement problem isn't a problem (Copenhagen, some others).
(2) Quantum decoherence solves the measurement problem (MWI and several others).
And what does quantum decoherence have to do with the measurement problem? It shows that through interactions with the environment, a system in a superposition evolves one that approximates a mixed state "for all practical purposes." In other words, interactions with the environment break a quantum superposition into a mixed state, i.e. a classical probability distribution.

Reminder: classical superpositions are always indistinguishable from mixed states. The main issue is with the quantum measurement problem is actually how it deals with quantum superpositions. That's what Schrödinger was talking about in his thought experiment in the first place! Therefore, I don't think I'm being at all silly when I say that the essential difference between CM and QM is their treatment of superpositions, and not entanglement, and not state collapse during measurement.

[jwl]

As crazy as that claim is - they don't really need to back it up, assuming they're actually capable of experimentally reproducing what they say they've accomplished. Quantum entanglement is inherently a phenomenon which acts at arbitrary distances, instantaneously. 3 meters or 3 light-years, it doesn't matter. Yes, it violates relativity, yes it's fucked up, yes, Einstein didn't like it. But subsequent experiments have empirically verified that it does, for whatever reason, actually happen - somehow, the state of one particle can be instantaneously affected by measuring the state of another particle, even if the two particles are on opposite sides of the galaxy (or even Universe.) How the actual information is transmitted is unknown.

So you basically just ignored my post to repeat the exact thing I was criticizing?

WHY do they not need to back it up? That's ridiculous. This is science. You can't make statements without some way to back them up. What about the ability to transmit this information over 3 meters implies that 3 light-years or any other arbitrary distance is possible? Yes, it may be possible in theory, but why can they just say that distance is completely meaningless when they have not actually done any experiments to verify that this is the case? This is exactly the problem I was complaining about. There is no venerability or accountability in such a theory, which makes it empirically pointless. What work has been done that implies quantum entanglement completely ignores distance, as implied? Just spouting nonsense like "oh, well, in theory it does" is just that - nonsense. That isn't how fucking science works.

This is like me pricking my finger with a pin, noting I feel the sensation instantly, then saying in theory nerves allow for instantaneous transfer of information over any distance.

The theory is entirely verifiable (assuming Channel72's interpretation of the results is correct). You can test it at X light years to see if it is true. But that doesn't mean you can't assume the theory is still true until you see reasons to believe otherwise. The whole nerve idea would work fine until it was found nerves are not instantaneous over any distance at all.

If you're correct, the Schrödinger's cat thought experiment doesn't even require a Geiger counter. We could just hook up some classical probability distribution apparatus to the hammer and poison flask, like a mechanical coin flipping device, or an ordinary computer that uses some internal source of entropy to produce a (pseudo) random number - and we'd have the same counter-intuitive problem of a dead/alive cat before it is observed.

But of course this is completely silly because nobody truly believes the collapse of a classical probability distribution is non-deterministic. The outcome of a coin flip, after all, depends on things like air pressure, gravitational pull, wind factor, the exact motion of the hand, etc. Even a Turing machine is theoretically incapable of producing a non-deterministic event, which is why truly random number generators rely on quantum phenomena.

This is pretty much right. The entire point of the Schrödinger's cat paradox is to demonstrate the difficulty in separating quantum systems (radioactive decay) with classical ones (cats). At some point, the flipping of a coin or a computer will end up coming back to a quantum process in theory, it's just the effects will (usually) be almost completely cancelled out before it gets to that level.