Two Big Steps Toward the Quantum Computer
popularmechanics.comThis is all I can think of when I hear about this stuff: http://upload.wikimedia.org/wikipedia/commons/4/4e/Eniac.jpg
> But here's the tricky part. The scientists can put the rubidium atom in superposition, so that it is simultaneously in that energetic state and not in the energetic state. It's on and off. Because of this, the photon both does and does not enter the mirror, mingle, and gain its polarization change. And the photon, by virtue of having both changed and not changed, carries that superposition information and can bring it to a different atom-based qubit.
I know almost nothing about Quantum Mechanics, but this sounds amazingly ingenious.
So what did they do, some sort of parallel universe transistor where the rubidium atom acts as the gate? If you assemble a processor out of this, will it compute all possible computations at the same time? And, last but not least... how do you make it converge to the computation you actually want?
Quantum computers make my head spin.
Yeah, one interpretation of what a quantum computer does is that it performs all computations in parallel. However... you run into problems when you try to read the output of this computer. In order to read the output you must measure a quantum state, and when this happens you only get to see one answer at random, not all of them. And that's not really useful. But if you can design an algorithm so that all of the parallel computations add together to form one answer (i.e., get all the wrong answer's probabilities to cancel out), you can get an exponential speed up. The key takeway is that quantum computers are not faster for general problems - quantum computers are only faster for problems where a special algorithm exists (like Shor's algorithm for factoring).
I was under the Belief that we currently have no way of adding the qbits together. We can set a qbit and read it but we can't currently get the particles in the quantum state to interact.
The article however paints another picture.
Some groups have even been running Shor's algorithm on multi-qubit computers. https://en.wikipedia.org/wiki/Quantum_computer#Developments It requires the qubits to be entangled to do the computation, and in the later ones (not IBM's 2001 work) the researchers did observe entanglement.
Technically, you can run Shor's algorithm on non quantum computers, with an exponential speed up (but fully within capabilities of modern computing vs. small quantities of qubits)
Here is a Google tech talk with Eric Ladizinsky that got posted a few days ago. https://www.youtube.com/watch?v=eIEy1KHk0rk
Very interesting listen if you have the time
IANAQP, but doesn't this article describe measuring the state of a quant? Shouldn't measuring put a quant into a defined state, i.e. destroying its superposition?
Measuring certainly would, but I don't think that's what it's describing. Instead, I think it is describing connecting the qbits together and allowing them all to share the same super-position through the entangled photon.
Am I missing something? Isn't a quantum computer just a non deterministic computation mechanism?
It feels wrong to suppose that answers come out in an "instant", or a mere one step.
Continuum of Singularity, so to speak.
When two network quantum chips can communicate instantly (not at speed of light--instantly) over infinite distances, using entangled atoms, then we can ditch the tecos.
I'm so looking forward to that.
Faster-than-light communication using entanglement is not possible, according to the standard rules of quantum mechanics.
(Dropping some credentials doesn't seem inappropriate here: I worked professionally as a researcher on quantum computing and quantum information for 13 years.)
Quantum mechanics doesn't allow instantaneous communication.
QM does allow some additional forms of spontaneous coordination, like in quantum pseudo telepathy games [1], but there is no communication. There is no back-and-forth decision making, just after-the-fact correspondence.
Unfortunately, the distinction between classical communication and quantum coordination is kind of hard to explain. It has to do with the difference between stochastic and unitary matrices [2].
1: http://en.wikipedia.org/wiki/Quantum_pseudo-telepathy 2: http://en.wikipedia.org/wiki/Unitary_matrix
I'll bite the bullet and try to explain it in layman's terms (with not attempt at rigor):
When you measure particle A, something happens to particle B. Unfortunately, particle B always has 2 potential outcomes (let's say, with 50% chance of being RED and 50% chance of being BLUE). So when you measure B, you find "B is red", or "B is blue".
Now when you measure particle A, imagine you change the probabilities for B remotely to 90%/10%.
But when you're the guy at B, and you machine say "RED!" ... did that just happen because A changed the likelyhood, or did it happen because, well, there was a 50/50 chance of it happening?
It is more subtle than that, but that's the gist of why you can't send information. (yes when you repeat the experiment and A and B compare their results, the probabilities have changed, but for a single measurement you never know if you got it by chance or if you got it because A did something).
>When you measure particle A, something happens to particle B.
Couldn't that be the signal? For example, one person could tell the other: "When B resolves, press the button!" It wouldn't matter if B resolved to red or blue.
Is it that we cannot detect whether B is in a superposition state without observing it and therefore resolving its state to one 'position' or the other?
(I hope I'm not the only one on HN with an incomplete understanding of current quantum theory.)
What do you mean "When B resolves, press the button"? "When you measure A, something happens to B" ... that's the problem, we can't know what happend. Only after the fact, and after repeating the experiment and sharing the results between the two parties.
You're Alice, and you want to send Bob the message "1001010". Let's start with the first "1". You measure A and see "red", and thus alter the probabilities of B to "90/10" ... and you think to yourself: Awesome, I just sent a "90/10" probability to Bob, and that means "1". If I had gotten "blue", Bob would be receiving a "50/50" probability.
Now, you're Bob at Alpha Centauri, and a particle arrives. Then what? No matter if you get "red" or "blue", you'll never know if it happened as 50/50 (the inherent randomness of any quantum measurement), or because of the "90/10" probability. So when you have to write down was it a "1" or a "0" ... you can't know.
At that specific moment, in your lab, when particle B arrives .... the result doesn't tell you anything.
Once you meet again, or send an email (you can compare your stats and find out that, statistically A affected B ... but if you need email (classical communication) to find out, then it's definitely not faster than light)
There are more subtleties about the uncertainty principle, orthogonal basis, etc ... but you would need a more formal language to express it.
I understand what you are saying but that's not what I meant to ask. The signal I'm proposing is not red/blue, it is superposition/resolved. Maybe this is more clear:
1) On Earth, Alice creates two entangled particles, both in superpositions, and gives one to Bob. She tells Bob: 'If your particle ever loses its superposition and resolves, whether to red or to blue, press the button!'
2) Bob goes to Alpha Centuri, Alice remains on Earth.
3) Alice wants Bob to press the button. She does something to resolve her particle.
4) Instantly, Bob's particle loses its superposition and also resolves. Bob gets the message and presses the button, in much less time than 4 years.
Why wouldn't that work? I suspect because you can't determine whether or not the particle is in a superposition, but my understanding is limited. Maybe the basis of my question is wrong.
You have to know the "phase" at which to measure the quibit, which you would need to communicate through other means. Here's some more info from a class a took a couple years ago. http://www-inst.eecs.berkeley.edu/~cs191/sp12/notes/chap1&2....
Chances are FTL communication is not possible.
http://curious.astro.cornell.edu/question.php?number=612
http://en.wikipedia.org/wiki/Superluminal_communication
The book, "Why E=mc^2" is an exceptional read. If I recall correctly, it explains why light-speed is a universal limit:
http://www.amazon.co.uk/Why-Does-mc2-Brian-Cox/dp/0306819112
Stating whether FTL communication is possible at all may require hedging. Stating that it can't occur over quantum superposition does not, as I understand it. The math is clear; there's no information traveling. It does not matter how sophisticated our quantum computers get, there will not be any FTL communication coming out of them.
If we could do that at all, we'd almost certainly be able to to it today, anyhow. We have multi-q-bit QM computers, they just aren't of a practical size for computation. But if FTL communication was possible, it would have been done with them already.
Mind you, this would imply that the telcos probably can't send data FTL either, so it's all good. :)
If at any point in the future of human history FTL travel or communications gets invented, we would already know about it.
As far as I understand, faster-than-light communication violates causality in some reference frames but not in all. FTL communication without arbitrary tims travel might be possiblem
how would you entangle the photons in the first place? from what i understand they have to be in close proximity or even emitted from the same source. once entangled, they need to be maintained in this state very very carefully. i think a few hours is the record currently.
to really reap the benefits of entanglement for anything other than microsecond HFT algos, like for human telecommunication, you would need to physically separate these photons by hundreds of thousands to millions of miles. i dont think you can keep them entangled while transporting them that kind of distance.
i could be really really wrong on all of this :)
>> i dont think you can keep them entangled while transporting them that kind of distance.
The main reason for this is decoherence (the pair of entangled photons interact with other photons, or particles along the way) ... and lose their connection because they have to share it with the other particles [1].
People have come up with clever tricks to fight that. It's called distillation of entanglement [2]. You share 1000 pairs of particles, all weakly entangled because they have travelled a long distance and rubbed with the wrong particles along the way. Then each party at each end combines the particles with some measurements and classical communications (phone or internet) and ends up with a single pair, far away, highly entangled.
>>to really reap the benefits of entanglement for anything other than microsecond HFT algos
These systems are high latency as they often are supplemented by classical communication and post processing. They are used by banks (in Switzerland) to do secure point-to-point communication. But they only protect the channel and not the copy of Windows 2000 running on either end.
Yeah, not saying this was possible now. But in 10, 20 years?
Perhaps they can be entangled at the factory in the "special room" then kept stable over the next 10 years. That would be enough, no?
Sure. Last year, they stopped a photon for a solid minute. http://www.extremetech.com/extreme/162289-light-stopped-comp... Note this preserves the quantum state of the photon - it's not just storing the energy and then re-emitting a different photon.
It is my limited understanding that the idea of "a different photon" would be sort of irrelevant here as quantum state is the only thing that differentiates objects such as photons from one another. Outside of that, photons are essentially fungible. I don't really understand the finer points of the experiment however and I could be totally wrong in my understanding.
I can imagine a world where they're more than just fungible-
They could even be literally the same object. See http://en.wikipedia.org/wiki/One-electron_universe
Oh yeah, I think the current model is that it gets stuck in an electron somehow. But since it preserves the quantum state, it's still usable for quantum crypto and quantum computing.