An Animation About Quantum Computers

http://quantumfrontiers.com/2013/08/22/the-most-awesome-animation-about-quantum-computers-you-will-ever-see/#content

by jonbaer • 13 years ago

84 44 13 years ago

quantumfrontiers.com

bhauer• 13 years ago

As soon as I saw the page, I thought, "Is this a Jorge Cham [1] project?" Sure enough, it is! Awesome.

[1] http://www.phdcomics.com/

fatjokes• 13 years ago

Jorge Cham is my hero. What's the most logical step after getting a mechanical engineering PhD from Stanford and a post-doc at Caltech? Become a cartoonist, of course.

fc2• 13 years ago

Probably a dumb question, but how can we know for sure that it's intrinsic randomness? Maybe we just haven't figured how it really works?

olvar• 13 years ago

Not dumb at all! Is a discussion that can be a little misleading, so be wary and don't blindly trust some random guy on a forum.

In the video they relate this randomness to the fact that qubits live in a superposition of states, in contrast to bits, which can be either 1 or 0, but not both. This means that when you measure a qubit, you would still obtain 1 OR 0, but you will obtain each with certain probability. So, how is this different to just picking a random bit and then measuring it? You could fabricate a way of picking those classical bit that yield the same distribution than the qubit.

We find a difference in one of the very first experiments that highlighted the quantum nature of the microscopic world; the double slit experiment: you have two screens, one with 2 small holes and the other is placed behind that. Then you send a bunch of classical particles through the punctured screen and you receive the counts on the screen behind it. Then you have a distribution of places where those particles hit. Then you do the same with quantum particles this time. You obtain a different pattern (this is important, is different to the classical pattern). This pattern is called "diffraction" pattern and means there was a wave like interaction between the particles. So now you try to do the same experiment with quantum particles, but sending them one by one, so no classical-like interaction could exist. Amazingly, you find the same diffraction pattern. This means one particle can "interact" with itself. The way physicists explain this is by allowing quantum states to exists on this superposition states which induces intrinsic randomness due to the fact that when measured, each particle could only have travelled through one of the slits, or each qubit (after measurement) can only be 1 OR 0, but not both.

This difference between classical and quantum description of nature goes deeper, and makes mechanisms as entanglement and teleportation possible. "Bell inequalities" are a great example of of some other, related ways in which both differ.

IanCal• 13 years ago

It's not a dumb question, that's what we should start by assuming because it makes a lot more sense. This is referred to as "local hidden variables".

Bell's theorem (http://en.wikipedia.org/wiki/Bell's_theorem) and the tests around it show this to be false, things just don't act like there's a variable we simply don't know.

HCIdivision172• 13 years ago

That is exactly the experiment I couldn't remember the name of and tried to ramble my way around it via another route.

This is the clearest way to understand that it must be truly random and not just hidden ignorance of state, and has been tested to a remarkable degree of accuracy.

HCIdivision172• 13 years ago

Not knowing the specific state - and being incapable of knowing - is the critical bit.

TL;DR: (edit) ignore the rest of this and just go read a lot on Bell's Theorem and the Double Slit Experiment. Both demonstrate concretely the essence of the weirdness.

(I'll leave the rest below just to keep me honest about my ignorance.) it's a provably special state where randomness allows it to have the needed characteristics for the math to work at all. [1]

Also disclaimer: I'm not a PhD physicist, just a BS in engineering physics. This is merely how I cope with the idea.

Think of it like this: given time, chaotic systems become unpredictable and noisy. An electron whizzing around an atom doesn't need much time before it's state is utterly unpredictable at any measurable level (Heisenberg Uncertainty assures this must happen eventually, and pretty fast it turns out).

So what if we isolate that little electron from the outside world? Then even the universe becomes uncertain, and must assume ANY possible state is valid and happening at once, blurring them together based on how likely each is given the last known observation! Otherwise the next observation could be contradictory to expectations, and seg fault the universe (perhaps just locally). This way though, the universe can always say "well I thought it could have been that" and be correct, even if it only thought it really unlikely.

But until the universe interacts with the state then, once coherence [2] sets in only probability can hint at what state it is in. Thus it is random in the before-the-fact sense like the probability before rolling a die, as opposed to an after-the-fact sense like a coin that is already flipped yet hidden.

And - here's the catch - the math needs to treat it as unknown. To label the parts of an unlabeled system will break the algebra! And if a distribution of states is unknown in a mathematically precise ignorance, this is indistinguishable from true randomness. Such pure conceptual states are rare in our messy universe (at least observationally), but they do exist and can be proven to exist in the same way a one time pad can be proven to be cryptographically secure.

Thanks for being my duck, I needed to think about something interesting today!

[1] Perhaps the most amazing bit is that if it weren't random, then it would act differently, since the distribution of observations would vary from the expected random field after enough observations. And this has been done! Study the double slit experiment deeply, as it really is a fantastic empirical example of why interfering via observation adjusts our interpretation of the results. [2]This term is jargon here, like quark colors. It's not coherence like a laser's.

Xcelerate2• 13 years ago

but how can we know for sure that it's intrinsic randomness?

We can't be sure (even though Luboš Motl disagrees). The universe may not be intrinsically random if superdeterminism is true. However, this isn't a popular idea in physics; in fact Gerard t'Hooft is the only Nobel prize winner I can think of that supports this idea.

Local hidden variable theories have almost all been ruled out by experiment in the last few decades (the same kind of theory Einstein was trying to find). There's still some loopholes, but, eh...

Colbeck and Renner have written a pretty powerful paper that given the assumption that measurements can be chosen freely, no extension of QM can have improved predictive power than the current theory of QM, regardless of whether this randomness "comes" from somewhere. Here's the paper: http://arxiv.org/abs/1005.5173

Strilanc2• 13 years ago

Pragmatically, we're kind of forced to assume simplicity until we can show otherwise. "Maybe we just haven't figured it out" applies to everything. Maybe we just haven't figured out how to violate energy conservation, or time travel, or escape the matrix.

Quantum randomness has a quality that makes it hard to explain in other ways. The results of measurements can't be made up on the spot, because they need to correlate with remote measurements. They also can't be stored ahead of time, because of how the correlation depends on which measurement is performed.

abritishguy• 13 years ago

The title is too bold, after seeing a claim like that the video is disappointing (doesn't really explain much just states things).

The video is very good and interesting but certainly not "The most awesome animation i will ever see". Upvote video, downvote title.

elboru• 13 years ago

I'll oversimplify my question, how will be possible for us to get useful information from a system based in randomness?

For my understanding, I can ask a quantum system: is this a cow? and it'll answer:

1. Yes

2. No

3. 72

libraryatnight2• 13 years ago

This was my question too, I don't understand how you can begin to make a computer out of something with - as they explained - intrinsic randomness. And I didn't quite understand, how, if we can't see or ask the computer how it did something, how we can be certain of its answer or know that we've built something that does what we want.

takeoutweight4• 13 years ago

You can know the probability that the algorithm will output the correct answer. If the probability of a correct response is > 50%, all you have to do is run the algorithm multiple times to get within, say, 99.99% certainty that the most-commonly reported answer is correct.

geuis• 13 years ago

The video really doesn't do a good job of explaining how a quantum computer can work. They just hit the basic high points because this video is intended for a general rather than specific audience.

A qubit exists in multiple states until measured. When you entangle multiple qubits together, each one increases the power of the system exponentially. Quantum computers are only good for certain kinds of problems like factoring large prime numbers. Luckily quantum systems are good at things classic computers aren't, and vice versa. You have to state your "question" in such a way that it causes the overall state of the quantum system to resolve on your answer.

I know my answer isn't much better. I only have a high level understanding with none of the math behind it. There's good videos if you search and watch a few times to get the concepts.

I can't remember the exact scenario that was explained to me, but it was something like this. Imagine the Traveling Salesman problem. You want the best route. But with a regular computer, it might take thousands of years to calculate the best route. With a quantum computer, if you structure the measurement correctly (this is how you "program" a quantum computer) then only the best route is what the system decoheres to.

There was also something about light and how if you had the right filter, only the answer gets through. I really forgot the details of that one, but maybe it'll help.

flebron• 13 years ago

Pfft, anyone can factor large prime numbers :)

ssivark6• 13 years ago

To substantiate your point: Essentially, quantum computing is useful for problems that could take help from a large degree of parallelization. Quantum mechanics has an inherent parallelization at it heart. Otoh, simulating that amount of parallelization with classical resources will be exponentially complex.

Strilanc2• 13 years ago

Quantum computing is sortof like... destructively interfering all the non-solutions so the only likely measurement results are solutions.

The state of the computer is a vector of complex amplitudes (one entry per possible answer, n qubits -> 2^n possible answers). The operations of the computer correspond to unitary matrix multiplications against the state vector. An entry's probability of being the result is the squared length of its final amplitude.

I try to explain it better here: http://twistedoakstudios.com/blog/Post2644_grovers-quantum-s...

valtron• 13 years ago

The visualization for complex product is amazing!

ssivark6• 13 years ago

Haven't seen the video yet, but I'll take a shot at your question. To my knowledge, the original motivation of a "quantum computer" was to simulate quatum mechanical systems (leading back to Feynman's lectures on computation). Simulating quantum (or other kinds of random) events with classical resources (conventional computers) is a hard problem in the sense that it's exponentially more complex than a classical problem.

Conversely, a quatum computer is very useful to solve such problems, think of: genetic algorithms, Monte-Carlo, etc. It is expected to do way better than classical computers. That doesn't necessarily make the simple problems easy on a quantum computer. Asking it the sort of question you did is not a measure of it's usefulness.

Since I'm not an expert on this, my aswer might be a little loose, but that's the basic idea.

FrankenPC3• 13 years ago

Anyone know the answer to this question? OK...so, if multiple wave functions which are overlapped make up the concept of super-position, and we can measure the wave function at any given time, why can't we take enough measurements to formulate a wave theory that will predict where the wave will be at ANY given time?

qwerty_asdf2• 13 years ago

We can predict the behavior of a continuous signal, since it's an ongoing stream of similar events. And we can predict the averages that we can expect to see, but for a single, individual emission of one photon, when we provoke or stimulate the emission of that photon we can't be sure EXACTLY how the photon was emitted, until it bombards the detector, and changes the detectors state. THEN we can infer from the detector, the original state.

The reason it's difficult to know the precise situation of a photon, in-flight, is because it's been emitted from a moving target, whose precise state is difficult to control perfectly: a live atom.

Without having perfect control of the atom that sends off the spinning, oscillating particle, we have a hard time discerning the exact circumtances of the spin/oscillation imparted upon it. We can hazard guesses, but when we're right, it's only because we get lucky. Not because we controlled it.

EDIT: The reason is there are two factors driving the emission:

1. The external behavior of the electon shells

2. The internal nuclear behavior of the protons and neutrons (which do not behave simply as a clump of spherical marbles, the nucleus is it's own volatile, nebulous construct)

And, oh yeah... 3. Any other random cosmological events that might sail in from outer space (literally), as fortuitous signals

FrankenPC3• 13 years ago

Well, that makes the most sense to me. But, I've always thought this is how the universe worked. Infinitesimal forces acting on a given outcome make precision prediction impossible. I still don't get how quantum mechanics is "spooky". Maybe without a math degree I never will.

Xcelerate2• 13 years ago

we can measure the wave function at any given time

We can't. Nobody knows how to measure a wavefunction. Instead, what you measure is one of the eigenvalues of a linear operator that acts on a wavefunction (which particular eigenvalue? You can't always predict this; hence the randomness in QM). For each "observable" you want to measure (position, momentum, spin), there is a specific linear operator associated with it. For instance, in position-space, the position operator is just the identity operator.

If you haven't taken linear algebra, I can clarify this further.

FrankenPC3• 13 years ago

I have a suspicion what you are trying to tell me is at the root of this. But, no I do not remember any complex math. Yes, I took multiple calculus classes in college, but I don't remember any of it. That was decades ago.

ArbitraryLimits• 13 years ago

You don't measure the wave function, you measure the state of the underlying system (qubits or whatever).

When a measurement is made, the wave function, which gives the probability of finding the system in any particular state, "collapses" so all the probability mass is on the state you just measured.

If you keep measuring the system quickly (relative to the time it would take to change significantly) you actually stop if from evolving the way it would have if you'd left it alone, as if you dropped a ball and kept staring at it and it dangled in the air instead of falling.

ORioN63• 13 years ago

The black box thing, really looks like a monad.

takeoutweight4• 13 years ago

You may be interested in the category-theoretic presentation of quantum theory. Turns out we're in more of a monoid than a monad! :)

http://math.ucr.edu/home/baez/quantum/

ssivark6• 13 years ago

Here's some recent work, trying to understand QM using notions from category theory -- http://arxiv.org/abs/1303.6917

ibudiallo• 13 years ago

Finally after watching this video, I can tell with great confidence I don't understand the first thing about quantum computing.

kghose• 13 years ago

I'm sorry, but this video did not make anything clear to me. I really liked the visuals but there was absolutely no point in the video that I could find. I feel the narrators could have done a better job explaining what quantum computation IS, rather than going on about superposition. Are these parallel computations going on? What?

JTon• 13 years ago

Unfortunately, I was lost as well. I have no background in quantum mechanics. I can't help but feel that was needed as a prerequisite to appreciating this video. Sort of like a beautiful analogy. It falls part of you don't understand the concept to begin with.

ssivark6• 13 years ago

Unfortunately, that video makes a lot more sense only if you understand quantum mechanics or quantum computation.

The video doesn't realy describe what quantum computation is but it rather describes what feature of QM makes quantum computers different from classical computers (superposition and quantum correlations) and why is it so difficult to build or use a quantum computer (decoherence, and the fact that you can't observe or clone the "registers" at intermediate stages in a computation).

The video addresses the notion of quantum computing quite obliquely. I would expect this to be the first of a multi-part series where this is explored in more depth (especially given the attention it's receiving).

sophacles• 13 years ago

It is - well its part 2 of an N-part series. Apparently they are being released as they are being finished, the last one having come about about a month ago. The link is in the text a handful of pixels below the youtube embedding.

jeffwass• 13 years ago

Here's a java simulation I wrote about 10 yrs ago showing visually how a single quit behaves under various operations. Didn't quite get to completing the two quit case (which is where entanglement enters the picture and he's more interesting).

https://github.com/jeffwass/Squankum

shire• 13 years ago

This stuff just blows my mind. I really want to get into Quantum Computers but I don't think I'll be able to wrap my mind around it. It's really amazing how much stuff we don't know in this universe.

An Animation About Quantum Computers

Related Stories