Is Quantum Computing for Real?

[Quotidian]

This week, in Sydney, there was the announcement of not one but TWO Quantum Computing centres opening at two rival universities (UNSW and USYD). There was a feature on one of the current affairs shows here with the very charismatic guitar-playing professor of one of the centres, which has had a major funding injection by Microsoft. The other centre was opened by none other than the Prime Minister, Malcolm Turnbull. (Article is here.)

Now I understand the very basics of quantum computing - that it is based on QBits, not binary bits, which in turn exploit the phenomena of entanglement and superposition to perform calculations that would be impossible to execute in binary systems. I understand that David Deutsch is one of the main theorists behind the field, and that he is something of a science super-star (who also happens to believe in t he Many Worlds interpretation of quantum physics, which is somehow important to his theory of quantum computing.)

But I have also read that there is scepticism as to whether a working quantum computer can be built, because of the 'problem of de-coherence', which poses a huge technical challenge. There is a company, DWave, which Google has a stake in, which claims to have produced a working quantum computer, but others, notably a physicist called Scott Aaronson, are sceptical, saying that the output could have been produced by a common-or-garden variety super computer.

So is the quantum computer hype? Or is it real?

 

[GTOM]

Theoretically, what makes a quantum computer superior to a normal, or an analog computer?

 

[Hornbein]

Theoretically, what makes a quantum computer superior to a normal, or an analog computer?

It is able to solve problems with large numbers of dimensions. It could revolutionize chemistry, since molecules are quantum systems with possibly thousands of dimensions. A quantum computer could predict what such a system could do. Nowadays chemists synthesize the molecule and then see what its properties are. Very slow.

Non-quantum computers will never be able to do that.

As to everyday linear if-then-else computing, quantum computers are inferior and likely to remain so.

 

[phinds]

Theoretically, what makes a quantum computer superior to a normal, or an analog computer?

They can do things in cryptography at speeds vastly above what will ever be even approachable by standard computers whether they are Von Neumann architecture (normal digital computers) or modified Von Neumann architechure (such as is common is DSP chips) and that are impossible in analog computers.

 

[phinds]

So is the quantum computer hype? Or is it real?

That is an excellent question and so far there is no clear-cut answer. There is a LOT of hype around them but there is also a lot of criticism and it remains to be seen whether they will fulfill their promise.

As hornbein noted, they will never be the be-all and end-all of computers since normal Von Neumann architectures will beat them ever time on many types of problems.

 

[f95toli]

There is no question that quantum computing is "real" in the sense that we know that it works and we can already solve (small) problems.
However, whether or not a "general purpose" QC will ever be of practical importance is still an open question.

However, what is somewhat more certain is that the technology for quantum simulations (which is a bit like old-time analog computers) will reach a point where it will be useful in just a few years. There are lots of problems that could be addressed using this technology, most notably in quantum chemistry and biology.
Another class of computer is the quantum annealer which is what D-Wave sells. These computers DO work and can already solve real-world problems but can't be use for e.g. cryptography/. However, they should excel for certain (important) optimization problems although so far they all circuits are so small that they can only be used for small(ish) problem where the solution times are about the same as you would get using a purpose built signal processor. But again, we should know in a just a few years whether or not they will be of any practical importance.

 

[cosmik debris]

Another class of computer is the quantum annealer which is what D-Wave sells. These computers DO work and can already solve real-world problems but can't be use for e.g. cryptography/. However, they should excel for certain (important) optimization problems although so far they all circuits are so small that they can only be used for small(ish) problem where the solution times are about the same as you would get using a purpose built signal processor. But again, we should know in a just a few years whether or not they will be of any practical importance.

The computers that D-Wave produce work but it is questionable as to whether they achieve Quantum speedup. I think that is where the controversy lies. Coherence times are short and it has been shown by some that a standard computer can produce results quicker than the D-Wave machine. There is quite a bit of argument between the two camps.

 

[rootone]

What programming language is used for quantum computing?

 

[phinds]

What programming language is used for quantum computing?

3 seconds on Google tells me that it's special "quantum languages". Dwave has their own.

 

[Fervent Freyja]

This week, in Sydney, there was the announcement of not one but TWO Quantum Computing centres opening at two rival universities (UNSW and USYD). There was a feature on one of the current affairs shows here with the very charismatic guitar-playing professor of one of the centres

Probably Andrea Morello from UNSW with the guitar. I vote for him to get it first.

I could watch that man talk about transistors and quantum tunneling all day -- look at those expressive, high energy hands...

 

[f95toli]

What programming language is used for quantum computing?

There is no special "language". So far everything is very low level, so you can use whatever language you want.that can create the appropriate sequence of pulses.
Most people will probably use Matlab or Python simply because they are the most common languages for this type of task.

D-Wave has written their own software interface for their hardware that can then be called from another programming language. I will try to remember to ask what is actually used next time I meet someone who uses it.

 

[FactChecker]

The current capability is certainly over-hyped, but the theoretical potential is not. If you have 500 Q-bits that can solve a problem in one step, that might solve a problem that has 2⁵⁰⁰ =~ 10¹⁵⁰ possible outcomes in one step. No traditional computer will ever be able to solve a problem that big that unless there is some shortcut algorithm to "branch-and-bound" it. Suppose there is a problem that does not allow a "branch-and-bound" shortcut solution. No traditional computer could solve it in a billion years. I don't know if there is such a problem, but it shows the theoretical potential power of the quantum computer.

Here is a possible example problem. It may not have any practical use. Suppose we have a function f(B), where B is a series of 500 bits. The value of f is 1 if B == B₀, a particular series of 500 bits, and f is 0 otherwise. We want to maximize f. It does not give you any other information about partial matches. Then there would not be any shortcut algorithm possible, and no traditional computer could find a perfect match to B₀. But, in theory, a quantum computer could.

EDIT (CORRECTION): I realize now that the example above will not work. In order to put that function into a quantum computer, you would need to know what B₀ is. If that were true, you could just read it.

 

[Tollendal]

Dear Quotidian,

The mathematical formalism adopted by Werner Heisenberg leaves clear that in the instant the location of the particle is made, all probabilities disappear. Strangely, since the formulation to this day, numerous discussions about the significance of this disappearance occur, maintaining that there is something misterious in it (Copenhagen interpretation). So, the problem isn't an engineering one, isn't about practical things. Quantum computers are theoretical impossibilities.

 

[.Scott]

There is basically DWave and everyone else. The DWave system performs a "quantum annealing" - it's basically a solution in search of a problem.

More generally, the goal is to be able to store and manipulate "qubits". A single qubit able to be in 2 states at the same time. In some devices, it can be in more than states than that. But a collection of qubits that has been deliberately manipulated by allowing that information in them to interact can be in a large number of states. For example, using simple binary numbering, 6 bits can code for any of first 17 prime numbers. But 6 qubits can code for all of those 17 primes - as a subset of the 64 possible combinations

Perhaps the most well-known quantum computing algorithm is "Shor's Algorithm", which factors large composite numbers - intriguing because the difficulty of such factoring is the basis for RSA encryption.

There are certain problems that are essentially impossible with normal computers that should be possible with quantum computers. But even once perfected, they will not be suitable for all computation problems. For example, the user interface to a Quantum computer will be a normal computer.

As far as how real they are, there have been some interesting breakthroughs recently. I would say that it is inevitable that quantum computers will be developed and prove useful within the next couple of decades - perhaps much sooner. And that is excluding the DWave device.

 

[phinds]

Quantum computers are theoretical impossibilities.

That does not follow from the rest of your post. How do you arrive at that conclusion? Do you have any sources or is that just a personal opinion?

 

[Tollendal]

Dear Phinds,

OK, if you think so. As you know, was formulated in 1926 by Erwin Schroedinger a partial differencial equation that describes how the quantic state of a physical system changes with time. For it, in 1933, he received the Nobel Prize (together with Paul Dirac).

It contains the factor Ψ, referred somewhat improperly as "wave function". The significance of it was not understood, until Max Born interpreted it as defining the probability of finding a particle in a determinate position of space. He received the Nobel Prize for it in 1932. The possibility can be represented by a Gauss curve, with maximum in the center and coming asymptotically to zero in the extremities. The mathematical formalism adopted leaves clear that in the instant the location of the particle is made, all probabilities disappear. Strangely, since the formulation to this day, numerous discussions about the significance of this disappearance occur, maintaining that there is something misterious in it (Copenhagen interpretation). Nevertheless, when we have a dice in hand before we throw it the possibility of each face falling upside is one to six. In the moment it falls upon the table and immobilize, to us it's clear one can no more speak of probabilities, as one of the faces was defined. Its obvious, there is nothing misterious in it, as even Einstein and Niels Bohr concurred.

Quantum computers are imaginated to work with the innumerable possibilities that would exist before the observation was made, as if they subsisted as physical things

It seems to me a supposed “observator's influence” is therefore nonsense.

It's what occurs when one believes that Physics necessarily must be described by mathematical formulas, even when they are not needed, as is the case. In this love for mistery, even today is frequent the understanding that the wave function signifies that the particle is in all places at the same time, and quantum theory would make possible the creation on a computer capable of realizing simultaneously infinite mathematical operations, a thing that would be useful, for instance, in breaking cryptographed texts.

Another common mistake that has the same origin consists in "multiple universes interpretation", that erroneously affirms the objective reality of the universal wave function, when it is a mere mathematical operation.

That I was thinking when I wrote, I would be glad if you agree.

 

[phinds]

I don't have the detailed knowledge to follow your logic, but I have to ask, why do you suppose that so many companies are paying a LOT of money to lots of really smart people to work on quantum computers if they are, as you say, a theoretical impossibility. That's what makes no sense to me.

 

[Tollendal]

Dear phinds,

I don't even imagine why so many grants are conceded, but I suppose reserchers are not really interested in Physics progress and have to pay their rent in the end of the month...

 

[phinds]

Dear phinds,

I don't even imagine why so many grants are conceded, but I suppose reserchers are not really interested in Physics progress and have to pay their rent in the end of the month...

That is EXTREMELY insulting to physicists. Do you seriously think that all those people are willing to waste their professional careers just to pay the bills? That's disgusting.

I find it MUCH more likely that you are missing something that they are not.

 

[FactChecker]

Dear Quotidian,

The mathematical formalism adopted by Werner Heisenberg leaves clear that in the instant the location of the particle is made, all probabilities disappear. Strangely, since the formulation to this day, numerous discussions about the significance of this disappearance occur, maintaining that there is something misterious in it (Copenhagen interpretation). So, the problem isn't an engineering one, isn't about practical things. Quantum computers are theoretical impossibilities.

This is too simplified. If there are several qubits entangled, their entire arrangement has a probability distribution and the entire thing becomes determined at once. In fact, that is what gives the quantum computer its potential. A system of 500 qubits could instantly settle into one of the 2⁵⁰⁰ ~ 3*10¹⁵⁰ possible combinations. It might give the answer to a problem that a conventional computer could never solve.. The trick is to make the quantum computer settle into a solution of a problem that we want to solve.

 

[Tollendal]

Dear phinds,

Yet someone as a professor highly considerated has posted in this same forum not far ago a declaration that said exactly this. I confess I was, as you, very much disapointed, but it could be an explanation.

 

[f95toli]

Dear phinds,

I don't even imagine why so many grants are conceded, but I suppose reserchers are not really interested in Physics progress and have to pay their rent in the end of the month...

Again, what you have written above is nonsense. The first demonstrations of (simple) quantum computing were done many years ago, so we do know that it works, The question is whether or not it will ever be practical, that is all.

Also, may I remind you that personal theories are not allowed on PF.

 

[f95toli]

There is basically DWave and everyone else. The DWave system performs a "quantum annealing" - it's basically a solution in search of a problem.
.

Not quite, If annealing works as hoped it would have quite a few practical applications; primarily in a fairly wide class of optimization problems in e.g. biology, machine learning and image recognition (which is why Google got interested). Also, D-Wave are not the only ones working on annealing,.

 

[GTOM]

This is too simplified. If there are several qubits entangled, their entire arrangement has a probability distribution and the entire thing becomes determined at once. In fact, that is what gives the quantum computer its potential. A system of 500 qubits could instantly settle into one of the 2⁵⁰⁰ ~ 3*10¹⁵⁰ possible combinations. It might give the answer to a problem that a conventional computer could never solve.. The trick is to make the quantum computer settle into a solution of a problem that we want to solve.

Great, how exactly that helps us? How do we know which combination will be the right one, unless we already know the answer, or already have an algorithm that can crack an encryption without try out everything?

It made sense to me, that we could model a molecule with it, but i still don't see, that just because a particle has a large set of opportunities, why it is way more efficient than some special architecture computer.

 

[FactChecker]

Great, how exactly that helps us? How do we know which combination will be the right one,

That is a issue that a lot of people are working on -- how to control the single-step process to solve the problem we want it to solve. The development and use of quantum computers is still a long way off.

 

[GTOM]

Again, what you have written above is nonsense. The first demonstrations of (simple) quantum computing were done many years ago, so we do know that it works, The question is whether or not it will ever be practical, that is all.

Also, may I remind you that personal theories are not allowed on PF.

They showed it works, did they show, that it is superior than a specialized architecture?
So there is an encryption like problem. Ok, the hundred particles can come up with the right combination in a second. Unless we use a regular computer to check the results, how do we know, they came up with the right solution? If another quantum computer checks the set of results, why don't they talk about its great clock rate instead of that huge number of possibilities?

 

[phinds]

They showed it works, did they show, that it is superior than a specialized architecture?

Yes, for some applications.

So there is an encryption like problem. Ok, the hundred particles can come up with the right combination in a second. Unless we use a regular computer to check the results, how do we know, they came up with the right solution?

Test encryption problems that are posed to quantum computers are ones for which the result is already know, so there is no need for any check on the answer. The point is to make sure that you get a known right answer so that you can have confidence in problems for which you do not already know the answer.

 

[FactChecker]

They showed it works, did they show, that it is superior than a specialized architecture?
So there is an encryption like problem. Ok, the hundred particles can come up with the right combination in a second. Unless we use a regular computer to check the results, how do we know, they came up with the right solution? If another quantum computer checks the set of results, why don't they talk about its great clock rate instead of that huge number of possibilities?

It is obvious when a message has been decrypted correctly. It turns into sentences. This is an example where finding the right solution is much, much harder than verifying that it is correct.

 

[.Scott]

Yesterday, IBM put one of its 5-qubit processors in the cloud.
You can access it here: http://www.research.ibm.com/quantum/

 

[Tollendal]

Dear f95toli,

I was chating friendly with other people and using somewhat loose wording, and I suppose that it could be tolerated in a relaxed discussion between friends. I beg your pardon if you could not accept this. Yet it seems to me what I said in such circunstances was clear enough in the
context: "quantum computers are imaginated to work with the innumerable possibilities that would exist before the observation was made, as if they subsisted as physical things" and "today is frequent the understanding that the wave function signifies that the particle is in all places at the same time, and quantum theory would make possible the creation on a computer capable of realizing simultaneously infinite mathematical operations".

Quantum computers may be feasible, I do not know, my interests in Physics are other, but the machines frequently announciated by the lay press that could make infinite mathematical operations at the same time, that evidently is not possible.

Evidently you are entitled to your opinion, but past week I received an encouraging message from someone as qualified as Professor David Chalmers, author of the acclamated book "The Conscious Mind: In Search of a Fundamental Theory": "Professor Tollendal, thank you, and all the best with your work!"

I need not to say I was proud of that!

Reference https://www.physicsforums.com/threads/is-quantum-computing-for-real.868729/


[URL='https://www.amazon.com/dp/0195117891/?tag=pfamazon01-20[/URL]

 

[FactChecker]

"today is frequent the understanding that the wave function signifies that the particle is in all places at the same time, and quantum theory would make possible the creation on a computer capable of realizing simultaneously infinite mathematical operations".

Quantum computers may be feasible, I do not know, my interests in Physics are other, but the machines frequently announciated by the lay press that could make infinite mathematical operations at the same time, that evidently is not possible.

As far as I can see, you are the only one here talking about infinite mathematical operations. IBM is allowing public access to a 5 qubit computer. D-Wave Systems has made a 1k+ qubit computer. (Many prefer to call it quantum annealing, not computing, but that is another discussion.) In any case, it is very possible to imagine the eventual development of a 100 qubit IBM-type computer. Suppose that computer can come to a single-step solution to a problem with 2¹⁰⁰ possible combinations to consider. That is not infinite. Far from it. Still, a conventional computer that can check 1 gig combinations per second would require 4 trillion years to try all combinations. With average random luck, it might find the solution in 2 trillion years.

 

[GTOM]

It is obvious when a message has been decrypted correctly. It turns into sentences. This is an example where finding the right solution is much, much harder than verifying that it is correct.

Obvious if there is something that reads it. Just because something can enumerate 2¹⁰⁰combinations in a second, that doesn't mean, that something can check all theese solutions.
Of course if there are local minimums in a function, so it is not just random, but rather like a n dimension gradient, i can imagine that those particles reach some ground level and stay there.

 

[FactChecker]

Obvious if there is something that reads it. Just because something can enumerate 2¹⁰⁰combinations in a second, that doesn't mean, that something can check all theese solutions.

No. Sorry if I was not clear. It picks one out of 2¹⁰⁰ possibilities as the solution. There is only one combination that has to be checked. And checking that one is easy.

Of course if there are local minimums in a function, so it is not just random, but rather like a n dimension gradient, i can imagine that those particles reach some ground level and stay there.

I assume you are talking about quantum annealing. Aren't local minimums at a high energy level less likely to be confused as a global minimum?

 

[GTOM]

As far as I can see, you are the only one here talking about infinite mathematical operations. IBM is allowing public access to a 5 qubit computer. D-Wave Systems has made a 1k+ qubit computer. (Many prefer to call it quantum annealing, not computing, but that is another discussion.) In any case, it is very possible to imagine the eventual development of a 100 qubit IBM-type computer. Suppose that computer can come to a single-step solution to a problem with 2¹⁰⁰ possible combinations to consider. That is not infinite. Far from it. Still, a conventional computer that can check 1 gig combinations per second would require 4 trillion years to try all combinations. With average random luck, it might find the solution in 2 trillion years.

That is not infinite, only practically infinite...
Ok it that can run through 2¹⁰⁰ combinations in a second. How do you tell the particles, stop when you found something meaningful to US, and don't proceed to a bad combination? Those particles only understand things like spin and ground energy level.

 

[FactChecker]

That is not infinite, only practically infinite...
Ok it that can run through 2¹⁰⁰ combinations in a second. How do you tell the particles, stop when you found something meaningful to US, and don't proceed to a bad combination? Those particles only understand things like spin and ground energy level.

The computer has to be programmed so that the state it settles into is the solution of the problem. It's not really "running through" the possible combinations. The program must be set up to influence the computer so that it will settle directly into the right solution out of all the 2¹⁰⁰ possibilities.