Friday, February 5, 2021

Qubit is one and zero at the same time

Comment:
quotes about quantum computing made in Finnish TV news watched by ~1.5 million people:

“Qubits are the quantum computing equivalent of normal computers bits, which are ones and zeros. Unlike a bit, qubit is not just one or zero, instead it can be both one and zero at the same time.”

“-This is a device that makes the impossible possible. Even if all supercomputers of the world would do something together for years, they could not solve things that quantum computer will at some point solve in a snap of fingers.”

“-Here we are learning how to build a quantum computer, learn to develop different parts, learn to write software and to apply them for different tasks. The real benefits will certainly come in the next phases at the end of 2020’s.”

The context of this was the announcement of 20 million euro funding to build Finland’s first quantum computer.

The commenter thinks that this is a distortion, but how would any non-physicist think anything else?

Wikipedia explains:

In quantum mechanics, Schrödinger's cat is a thought experiment that illustrates an apparent paradox of quantum superposition. In the thought experiment, a hypothetical cat may be considered simultaneously both alive and dead as a result of being linked to a random subatomic event that may or may not occur.
So a cat can be alive and dead at the same time. If you believe that, then a qubit can be one and zero at the same time.

If you think this is nutty, and I do, then get the Physics popularizers to stop saying that a cat can be simultaneously alive and dead.

The story also over-hypes what quantum computers will do for us, but again, it is just parroting what physicists say. In June 2019 I noted that experts were saying that quantum computing capabilities were growing doubly exponentially. If that were really true, then quantum computers would be doing useful things by now. But that could be decades away, or never happen.

7 comments:

  1. The whole 'The cat is alive and dead argument' falls down in so many ways. The entire conceit is just a constructed logical paradox like 'Can all powerful Zeus make a rock so heavy he can't pick it up?'. Paradoxes only serve one useful purpose, they test the ability of the listener to carefully unravel the flaw in what they have been told.

    First things first, who determines the awareness of the cat's state of life or death? The magical all knowing narrator who presents the problem? If one man does not know if the cat is alive or dead, and the other man does, which awareness determines the cat's state? First come first serve? Why would one observer be determinant and not the other? Is it a concerted groupthink effort of constipated quantum physicists staring at boxes with cats in them on the verge of death? If the man who didn't know the state of the cat also didn't know the other man who actually did know, would the cat be alive OR dead for one man, yet still alive AND dead for the other? This is important because it reveals the crux of the logical structure of the statement. The juxtaposition problem is only a problem relative to the observer, not the subject itself. This also leads to the other side of the problem as well, While your awareness of something may be uncertain, how can you then be absolutely certain someone else doesn't know? All this ignorance of cats being alive or dead, and no one bothers to ask if someone actually already knows.

    This would be kind of important as all the quantum buffoonery revolves around the premise of awareness as being the actual informer of reality. Seems to me if you have to admit your ignorance for the first case (is the cat dead or alive), you might as well acknowledge the second case (you don't really know if someone else knows either), in which case you might as well cut to the chase and say "I DON'T KNOW" instead of making up another universe to hide your ignorance in...along with the undead cat.

    Not knowing something has no influence whatsoever on the state of the 'something' you don't know. Why would it? Magic? If you don't know something, you somehow cause it to be indeterminate? Wow. This is just blatant intellectual narcissism.

    This is tantamount to saying that if the blind man doesn't see the door is open OR closed, then door is open AND closed at the same time...because he is ignorant. Bullshit. The blind man lacks awareness of the state of the door, that is all. It is not that the state of the door that is indeterminate whatsoever, it is the state of the blind man's awareness that indeterminate. The blind man is confused, not the door.

    Semantic bullshittery is not science. It's just poor logic and a cheesey sleight of hand waving.



    “Do not try and bend the spoon, that’s impossible. Instead, only try to realize the truth …
    There is no spoon. Then you’ll see that it is not the spoon that bends, it is only yourself.”

    -Spoon Boy

    a la The Matrix

    ReplyDelete
  2. When Sean M. Carroll says that energy is not conserved in quantum mechanics, he means that if you are ignorant about the energy level, and then measure it, the measured value might not match what you thought was likely before.

    ReplyDelete
  3. Roger,
    I've always had a problem with mediocre statistics by scientists who are more concerned about their advocacy than the truth. I fervently believe in GIGO. Many (who should know better) today want to believe that computers and math are magic, informed by their good intentions, and capable of wiping away uncertainty, and any size error bar they like with just a few ill informed variables in the name of 'the greater good'. If your initial values are garbage, your results are ALWAYS garbage, no exceptions, even more so if you are doing reiterative calculations in a computer simulation or model of some kind. If your initial values used in calculation are crap, so is anything that is produced from said values. Massaging made up numbers in hand tuned non-predictive models is not and never has been 'science', it's just another flavor of warmed over propaganda.

    We live in the age of priest-like scientists who want the power to make political decisions for us over our objections, and the right to squelch dissent or questioning their statements in the name of 'because I'm and expert and I said so'. When skepticism is absent, so is the truth.
    --

    I actually used to follow Carroll's blog, until I got sick of his smug technocratic political stances on everything, and his vapid attempts to redefine physics into useless metaphysical drivel, with himself as the chief pontiff.

    Sean Carroll strongly believes the baseline requirements for evidence in science are too restrictive for his enlightened opinions about using imaginary non-observable universes (which are nothing like ours) to explain how our own very observable universe works. Of all people you would think a self professed atheist and left-leaning scientist would know that invoking Creation ex nihilo ad infinitum miracles is not very scientific.

    Sean self identifies himself as a 'science communicator'.
    I just think he's just bullshitting for bucks.

    ReplyDelete
  4. This paradox began with Einstein.

    He took a classical object of an unexploded bomb, and then supposed that the occurrence of explosion was controlled by one quantum mechanical event: whether a radioactive atom decayed or not.

    In short, he took a quantum mechanical ket consisting of a linear superposition of two eigenstates (the exploded/not exploded basis). Then he logically connected it a classical system showing irreversible change, viz. the bomb. The pin of the bomb would necessarily get pulled if the decay occurred, but not otherwise. So, the pin and the bomb always are in one or the other classical states.

    Effectively, the pin should act as the input probe area of a QM measuring device, and the bomb as its needle. A rather fearsome measurement device.

    But then Einstein didn't stop there. He put this whole assembly inside a classical box, an isolated system in fact, just to make sure that the said quantum event was not subject to any influence from any other object.

    As to the connection between the classical pin of the bomb and the QM event of potentially decaying atom, he kept it vague, unspecified. In effect, he must have imagined the occurrence of the decay event to be subject to an in-principle un-specifiable kind of a randomness. (I don't know if stopped and thought about this direct implication of what he was proposing.)

    I think that Einstein had described such an arrangement in a letter to someone, probably to Schrodinger himself.

    Schrodinger then replaced the state of an un-exploded bomb with a live cat and that of the exploded bomb with a dead cat. I imagine that both of them must have been pretty amused both at the original arrangement and the improvisation effected to it.

    It is interesting to note that textbooks and physicists rather emphasize the linear superposition aspect, but they don't keep it confined to the decaying atom alone. Further, they also don't highlight the irreversible nature of the classical part of the arrangement, viz. the bomb (or the cat). None, I think, says that the bomb effectively acts as a measurement device. Instead, they choose to confound the paradox further, by strongly emphasizing a partial truth, namely, that the bomb (or the cat) can be regarded as a QM system in its own right. So, effectively, what they end up emphasizing is the decaying atom + the bomb (or the cat) system as being in a superposition of tensor product states involving the two parts.

    Effectively, this way of describing the thought experiment ends up deleting one crucial fact: Even when being described quantum mechanically, the bomb (or the cat) remains a part that can undergo an irreversible change to its own internal configuration. It also de-emphasizes another important (but somewhat less important) fact: The bomb (or the cat) therefore, effectively, acts a measurement device in its own right, even before the box is examined. Finally, as usual, the description also glosses over the aspect of preparation of the ket.

    If physicists themselves explain their specialization this way, we can't blame the mere pop-sci press to be any more accurate than they are. The blame lies almost wholly with physicists, and next to zero (or even to zero extent) with the pop-sci people of the kind mentioned in this post.

    Explaining a better way to verbalize these aspects of QM would involve making reference to a proper solution to the measurement problem (and to QM theory). However, keeping aside all explanations, I think the following might be satisfactory (contd in the next comment).


    Best,
    --Ajit

    ReplyDelete
  5. (contd from above)

    A classical system such as the register of a classical computer can only be *prepared* in one of the two states: '0' OR '1'. (The or here is the exclusive one.) Hence, the classical system can only be *measured* as *registering* either a '0' state OR a '1' state.

    Now, invoking other, more fundamental ontological assumptions, we can also say that, from the time the state was prepared, the classical system always *was* in a '0' state OR in a '1' state until measurement, and it remains in the same state during and after measurement too.

    In particular, a classical system cannot exist in a '0' AND '1' state at the same time. (By discretizing the analog states, we assign all of them to either 0 or 1, and no other possibilities.)

    A QM system such as a qubit can be prepared to be in a state that is a linear combination of a '0' state AND a '1' state at the same time. More precisely: \psi = c_0 |0> + c_1 |1>. Then, the QM system can only be *measured* as *registering* either a '0' state OR a '1' state, in any *single* act of measurement.

    Now, if you imagine a large number of identically prepared qubits and subject them to more or less identically conducted measurements, then, in the limit of infinite number of measurements, the fraction of times the measurement apparatus will register the '0' state will approch |c_0|^2, and that for registration of '1' will approach |c_1|^2. If c_1 = 0, only the state 'zero' is measured at all times, and c_0 = 0, only 'one' is measured at all times.

    Now, invoking other, more fundamental ontological assumptions, we can *also* say that before measurement, the QM system *was* in a superposition of both the '0' state AND the '1' state. Thus, before measurement both the states did exist at the same time, but only as fractions or parts of the entire state. The potentiality to measure either state was always there, until a measurement actually occurred.

    What happens after the act of measurement? Well, the mainstream QM talks only of the actuality of what's registered in the apparatus. In fact, following von Neumann, it "throws" the measured state "back onto the system", and says: The state of the system itself has collapsed to what was measured. Thus, if both 0 and 1 were superposed, and if 0 was measured, the mainstream QM says, the system itself (and not the instrument) got "reduced" to the '0' state. It refrains from talking anything about what happened to the unmeasured '1' state.

    In my new approach, I describe the physics of it differently. However, the maths of what events are observed at the detector(s) is the same as for the mainstream QM.

    Best,
    --Ajit
    (contd further)

    ReplyDelete
  6. (contd from above)

    Compressing it all for the truck driver or the school-kid who is listening it on the radio...:

    "Unlike a bit, a qubit is not just in the 'one' or 'zero' state, but instead, it can be put in a state that has both the 'one' and 'zero' states present in it, but only as parts, only as fractions of the total state.

    You prepare the QM system in this way. It then interacts with the instrument. The physics and mechanics works in such a way that the instrument can record only one of the two states at a time, but not both.

    That's like how an electrical switch that's perfectly midway can go to 'on' position, or to 'off' position, but not both.

    The perfectly mid-way switch is a classical system. We can't describe its being mid-way as also being at both ends: 'on' and 'off'.

    But a QM system is different. You can prepare and manipulate a state that is has both 'one' and 'zero' as parts, like how two waves can superpose. But at the QM level, the instrument can measure only one wave at a time, not both.

    Now, while the superposition state lasts, we can simultaneously manipulate both the parts in it, one and zero. Some advanced manipulations can speed up the computations.

    However, these superposition states are fragile, very fragile. They break down very easily, before any one gets any good chance to manipulate them for computations, or measuring them for recording the results.

    The fragility of maintaining the superposition states is the challenge in building a QC.

    It's a bit like taking an electrical switch in your hand and keeping it pressed perfectly mid-way while playing cricket---e.g., while tossing a perfect yorker with the other hand. In principle, always possible, but don't pin your hopes that someone is actually going to pull it off, ever---and certainly not by the end of 2020's.

    Best,
    --Ajit
    (Three part reply concluded with this one)

    ReplyDelete
  7. Dear Roger,

    Please allow me just one more, brief, reply, as an exception... (I won't explain anything; I will just state...)

    1. It would be much more enlightening to replace the cat (or even the bomb) by a simpler but classically describable part that can also trace one of two (or more) irreversible paths starting from the cusp-point (or, a part that shows a sensitive dependence on initial conditions). For instance, an electrical switch at its cusp, or a blob of a rigid-arm pendulum `resting' at the top-most point, etc.

    2. The mainstream QM models the universe as a linearly oscillating system.

    The only description possible with such a formalism, to describe the system consisting of {the decaying atom + the cat/bomb/electrical switch} is this: The total system only momentarily is in one of the two end-states (on/off); then it immediately "return"s from that end-state and spends some time in all the infinity of superposed states; then it momentarily touches the *other* end-state; then it immediately reverses its course... Ultimately, it keeps repeating back and forth (or the cat goes on becoming dead, then being in superposition of dead and alive, and then becoming alive again, repeating, in turns) an infinity of times.

    MSQM cannot model an irreversible change like an alive cat getting dead. Not at all. (No, it can't model a dead cat irreversibly becoming immortal forever either, for exactly the same reasons. This gets weird because life is involved. That's why I like the switch in place of the bomb or the cat.)

    3. While the decaying atom and the cat/bomb/switch must always form (superpositions of) tensor product states (with the atom and the switch as the two constituent parts for the composite system), if a classically describable change to the cat/bomb/switch is at all to be incorporated, there will have to be an infinitely large sequence of superpositions of (different) tensor-product states, and the system will traverse through such a sequence. Classical changes in configurations are continuous; QM doesn't mean that the physical space is quantized.

    MSQM then models the dynamics as following: The system traverses through all the possible dynamical branches (starting from the dynamical cusp) simultaneously, traces the dynamical path up to an end-state, then retuns, etc., all in a linearly oscillating manner.

    A proper explanation would model the system as passing only one way from the cusp to one of the two end-states.

    Neither would require a second universe, let alone an infinity of them.

    4. A better analogy would be for the cricketer to try balancing a fresh (unboiled) egg on his head as he pitches the perfect yorker (with the first bounce of the ball occurring precisely on a coin kept under the batsman's bat)---and, *also* with the egg remaining on the bowler's head in the end, unbroken. Yes, the egg may bounce up and down while the bowler is running, but it must keep returning to the top of his head and remain unbroken. Yes, presence of wind is to be assumed. Yes, such light rain as not leading to interruption of the game is also to be *allowed*.

    OK. Now, I am really done.

    (Won't write any reply here unless someone wants me to explain something from what I wrote....)

    Thanks and best,
    --Ajit

    ReplyDelete