All these years we had the buzzwords quantum and cloud separately. Today the buzzwords can finally merge to bring you quantum in the cloud. OMG, I predict the tech journalist will go crazy on this!!

A very similar quantum circuit simulator (in JS): http://www.davyw.com/quantum/?example=Grover%27s%20Algorithm

And now a shameless plug: If you want to learn quantum mechanics and quantum computing, you should check out my upcoming book on linear algebra: https://minireference.com/static/excerpts/noBSguide2LA_previ... available via https://gum.co/noBSLA

I have my own shameless plug to add.

Another drag-and-drop quantum circuit simulator that runs in your browser is "Quirk" [0].

Quirk is faster than Davy's simulator thanks to webgl. It calculates varying operations on up to 16 qubits in real time as you drag gates around. It also lets you drop inline state displays into the circuit. Plus you can link to circuits, though the links can be a bit long so you may want to do so indirectly [1].

That being said, Quirk lacks the ability to compile circuits into a gate. Also it doesn't have a QFT gate (yet). I'm working on it. [2].

0: http://algorithmicassertions.com/quirk

1: https://t.co/qi12H3aVU1 (6-qubit grover circuit with displays, near the getting-laggy limit in terms of op-count)

2: https://github.com/Strilanc/Quirk

That looks good. Yeah, mine wasn't really focused on performance so much as ease of use (being able to compile and reuse circuits).

One thing I never got around to that IBM did was measurement operators. They break my simple compilation strategy but would definitely make it more useful.

Oh! Didn't realize you would see this. Gives me the opportunity to say thanks for the simulator. I found it really useful for learning. So useful for sanity checks.

(Speaking of measurement. Quirk allows measurement... but it cheats. It refuses to let you hit measured qubits with operations that would superpose them again w.r.t. the computational basis. So the deferred measurement principle [1] applies, and there's no need to use density matrices and square the simulation cost.)

1: https://en.wikipedia.org/wiki/Deferred_Measurement_Principle

Ha! I feel obligated to click when I see "Try a Quantum Computer" in a title.

That's an interesting strategy. I'm going to have to dig deeper when I get a chance. You said you're also using WebGL for a performance?

PS: Those continuous gates are especially cool!

Yeah, using the GPU makes a huge difference. It's also incredibly finicky because the hardware and browser support varies so much.

Basically I represent an n-qubit superposition as a 2^(n/2) x 2^(n/2) texture with each pixel being an amplitude (with red=real, blue=imaginary components). Then I use fragment shaders to operate on all the pixels in parallel when applying a gate.

For example, here's the GLSL in the main method of the "UniversalNot" gate's shader [1]:

    vec2 xy = gl_FragCoord.xy - vec2(0.5, 0.5);
    float state = xy.y * outputWidth + xy.x;
    float hasBit = mod(floor(state / bit), 2.0);
    float partnerState = state + bit * (1.0 - 2.0 * hasBit);
    vec2 uv = uvFor(state);
    vec2 partnerUv = uvFor(partnerState);
    float control = texture2D(controlTexture, uv).x;
    vec2 val = vec4(texture2D(inputTexture, uv)).xy;
    vec2 partnerVal = vec4(texture2D(inputTexture, partnerUv)).xy;
    vec2 outUncontrolled = vec2(partnerVal.x, -partnerVal.y) * (1.0 - 2.0*hasBit);
    vec2 outVal = (1.0 - control) * val + control * outUncontrolled;
    gl_FragColor = vec4(outVal.x, outVal.y, 0.0, 0.0);

(The UniversalNot gate isn't possible in reality, but it's easy to implement in the simulator. So I have it implemented, but hidden away. You have to manually tweak the URL to contain a "__unstable__UniversalNot" gate to use it. But then you can use it for FTL communication shenanigans. [2])

1: https://github.com/Strilanc/Quirk/blob/master/src/circuit/Ga...

2: http://bit.ly/1roNMCE

Wow, this is awesome! The Block-sphere "spy probe" is an amazing way to show what's going on with each qubit[1]. I wish I had this when I was teaching the intro to QM course!

Suggestion: add some sort of menu for import/export and links to examples. I thought there was a limit to how long a URL hash can be, but it seems even very long ones will work.

[1] Of course, the local description doesn't really work for the Bell state (|00>+|11>): http://bit.ly/1rUFaos but still cool...

Thanks.

The other issue with the links is that the braces keep getting cut off by sites (as happened for your link, need to include that last } for it to work). But having the URL be the save mechanism is just so darn appealing to me.

I generally work around each qubit in an EPR pair appearing to be maximally mixed by conditioning the displays. Add a control to the other wire in the same column as the display, and you'll see the conditional state. Also there's the 2-qubit density matrix display.

I learned almost everything from chapter 1-9 including Fourier analysis at school. don't really know anything about chapter 10 (quantum mechanics). I work for amazon now so that the other buzzword cloud is taken care of.. Damn I was I chapter away from becoming a quantum cloud engineer.

Oh interesting, I will take a look at the sample book for a bit and see if this is something that I will buy. Been needing to relearn linear algebra for some time now.

Yes, LA is definitely worth relearning. I had a great time writing it since it was a chance for me to review a lot of material. Even more fun was writing about the numerous applications, in the end about 1/3 of the book is on applications[0].

Before you commit to reading 500 pages of intense math, you could probably get a lot of reviewing done with the short tutorial on LA[1].

[0] https://minireference.com/blog/linear-algebra-book-alpha-rel... [1] http://minireference.com/static/tutorials/linear_algebra_in_...

Will definitely check those out. Already bought the book. I like the writing style. Definitely easier to read than my college text books.

Perhaps I just had an odd copy, but I'd note that perhaps you might wish to have slightly darker ink for this book.

Did you get the print version from amazon? Or a bookstore? I've found the amazon print-on-demand quality to be stuck in "economode" sometimes. I personally recommend people to purchase from lulu because the print quality there is more consistent, though they too make mistakes sometimes.

Send me an email if the greyness bothers you—I'll be happy to send you a replacement copy from lulu, say when v6 comes out (no major changes; just more exercises because some people complained the onboarding was too hard for beginners).

> quantum in the cloud

You probably mean Quantum DevOps.

I'm looking for a technical cofounder with at least 10 years experience in Quantum DevOps

There is a 40% chance that you updated the code to the newest version.

Quantum ops you are never sure.

TL;DR: 5 qubits as a service. Interact with them on your tablet.

From the press release: "The quantum processor is composed of five superconducting qubits and is housed at the IBM T.J. Watson Research Center in New York. The five-qubit processor represents the latest advancement in IBM’s quantum architecture that can scale to larger quantum systems. It is the leading approach towards building a universal quantum computer." (http://www-03.ibm.com/press/us/en/pressrelease/49661.wss)

From the demo video: looks like you can use hadamard and CNOT gates as well as measurements. Somehow it sounds like you can also apply oracles (not sure how you specify what they are when it gets run). (https://youtu.be/pYD6bvKLI_c)

The oracle is just a circuit that negates one of the amplitudes. You specify it just like the rest of the circuit.

They use a trivial oracle circuit as an example. But Grover's algorithm works for any oracle circuit, so in principle it could be a circuit that verifies if its input is a proof of the Riemann hypothesis.

See also http://www.wired.com/2016/05/ibm-letting-anyone-play-quantum...

https://quantumexperience.mybluemix.net/ is the actual site

You've been able to do this sort of thing for some time, at least simulated, which provides the same results:

http://www.quantumplayground.net

Same results but faster! The IBM version runs slower than a classical simulation of it, making IBM's project a tech demo toy. (Obviously. If it were actually efficient, they would charge a fortune to use it.)

What factors contribute to the performance disparity, and where could efficiencies be found?

So these things presumably run software? Or is it basically a bunch of quantum bits controlled by a conventional machine? What is it exactly and what can you do with it?

On 5 qubits, you won't run much software.

And of course you want a lot of conventional machine around the quantum processor (even a hypothetical much bigger one), as you would around any special-purpose hardware (you have it even around a GPU).

Think of it as a coprocessor.

There used to be a separate FPU processor for math / floating point operations.

Then there was a separate GPU processor for graphics / parallel processing.

Now, this.

Close to your suggestion #2. There are electronics running software, plus a set of entangled qubits whose output probabilities are influenced by the machinery which the above mentioned software controls, and the software also interprets the output and decides when a sufficiently accurate answer has been found.

The particular configuration that the software defines for the controlling machinery (and how it sets up the qubit entanglement) is meant to implement the chosen quantum computing algorithms and to encode the chosen inputs.

For example, for a given instance of Shor's algorithm and for one given public key, you'd first calculate how to encode this configuration into the qubits, then load it up, to then run it over and over and interpret all of the outputs to look for a useful answer.

I've got no idea how the qubit programming is practically implemented.

But what am I going to do with it?

You could run Grover's algorithm on a database with 2^5=32 elements. https://en.wikipedia.org/wiki/Grover's_algorithm Or Shor's algorithm to factor a 5-bit number. https://en.wikipedia.org/wiki/Shor%27s_algorithm

Pinky: Gee, Brain, what do you want to do tonight?

Brain: The same thing we do every night, Pinky - try to take over the world!

______

Currently, I'm not aware of anything that's not able to be done with QC qubits that not able to be done with classical binary computation.

If you wanted, you can download quantum computer simulator, write some QC'ish code, run the code, and you'll never write any code that won't resolve with crashing or taking forever, though it'd be faster on the 5 qubits though.

I think David Deutsch came up with some simple algorithms back in the early 90's that are exponentially faster in a QC. I don't know if they implemented Hadamard/XOR gates in this thing.

1. Shor's algorithm

2. Grover's algorithm

3. Quantum simulation

If you're not doing those three things, throw it in the trash, more or less.

Is it possible to generate BitCoins on quantum computer?

Actually it is possible to bring the whole bitcoin system down with a (general purpose, large enough) quantum computer. The whole bitcoin system relies on ECDSA signatures, which can be broken with Shor's algorithm. You can't mine bitcoins faster with a quantum computer, because that's basically hashing, which you don't get a significant speedup. But you can steal everyone else's bitcoins.

You get quadratic speedup on mining, since the HashCash proof-of-work used in Bitcoin is a perfect application of Grover's quantum database search algorithm.

So while classical miners are brute forcing through a 2^70 search space, the quantum miner can find a solution in roughly sqrt(2^70) = 2^35 steps.

Other proof-of-work systems can be more quantum resistant, e.g. looking for a fixed-length cycle in a huge random graph, for which no efficient quantum algorithm is known.

Computing 2xSHA256 on a some hundred bytes block header with one section open for randomness will not be that easy on a quantum computer (the more complex the problem, the easier for noise to drown out the answer), not to mention you'll still need fast ASIC hardware to evaluate all QC outputs looking for valid blocks.

And you must reset part of the problem (the previous block hash, at minimum) when new blocks are released, adding some latency (you need to recompute the qubit configuration before resuming).

But perhaps a SIDH based proof-of-work algorithm could be implemented to further resist QC speedups. Don't know exactly how that would work. Does Grover's still apply?

You cannot steal bitcoins from an address that's never been spent, since you only know the hash of the public key.

There are about 2^256/2^160=2^96 possible full public keys mapping to the known key hash, so you could run Grover's algorithm to recover one in about sqrt(2^96)=2^48 steps, but given the slow cycle time of quantum computers, that's still going to be infeasible for a long time.

This is why address re-use is not recommended...

The quantum computer is going to suffer the same fate as plan 9. As Kernighan said, it's not as novel of an upgrade to warrant a mass adoption. Especially considering the price. Not everybody needs a real time black hole simulator...

i'm curious to know what the argument is for quantum computers doing supertasks.

For those interested in seeing the hardware behind this service, have a look at: https://www.youtube.com/watch?v=jf7D8snlsnQ

Quantum computing is one of those things I know I will never learn. It makes me feel like my current skillset has an expiration date, and I only have so much time before the bar for innovation is beyond my capabilities.

Most of the behavior of quantum circuits can be explained with just linear algebra. It's not that bad.

Plus, developers probably won't need to interact directly with this stuff when the day comes. I'm sure there will be some set of common quantum tasks abstracted away behind the scenes.

Will the computer only work when observed?

I'm not sure why I'd prefer this over Microsoft's LIQUi|>.

Edit: Okay I'm ridiculous, it has a real nice graphical interface.

What happens when you train a deep recurrent neural network on a quantum computer. I wonder if you get consciousness

You are getting downvoted, but no less than Sir Roger Penrose had the same idea for the origin of human intelligence.

And it was a bad idea when he said it, too.

Eh, the jury's still out.

I sort of agree with Scott Aaronson on why he thinks consciousness is finite[0].

The problem, IMO, that Penrose has is that his theories border closely on mysticism - there's some quantum "magical" process whereby consciousness arises. Both pictures are incomplete, but Aaronson's is the most plausible.

[0]http://www.scottaaronson.com/writings/finite.html

"This rule implies that 2, 17, and the number of possible human life experiences are all equally distant from infinity." [0]

Thanks for the link. I do have trouble with the above statement, and what follows it, as I would think that the number of states of the human body (or even just the brain & nervous system) would be infinite (that is, in the sense of the infinite cardinality of real numbers) and not countably infinite (as in the set of whole numbers). This would follow from the entire system operating as a biochemical process, and neuronal stimulus being an analog signal between a finite number of neuron cells. Thoughts?

[0] http://www.scottaaronson.com/writings/finite.html

Just so I understand you correctly, you're saying that the cardinality of the set of all possible states of neurons in one's brain is aleph-null?

Had to look that up again to be sure - no, instead what I'm suggesting is the number of possible states of neurons is infinite, and not countably infinite (implied by a cardinality of aleph-null).

I suggest the number of possible states is infinite since the signals between neurons are analog (though I acknowledge this one-word categorization is a major simplification of the actual process of neuronal stimulation).

Does finite imply non-quantum?

No, but there is no evidence that biological neural networks exhibit quantum effects.

Classical (Newtonian) mechanics seem sufficient enough to explain how neurons work.

No, of course not.

Can quantcom be used to compute genesis blockchain?

Hum...

Why does the guy who discovered the time of quantum decoherence (Claude Cohen Tanoudji) says quantum computing might be a scam?

Okay, people may think he is a fraud, but he was able to trap very few photons in a cavity and measure that quantum decoherence disappear with order of magintude of 10^-13seconds.

Ho! And maybe quantum computing is fast, but how fast is it for an operator to translate a NEW problem (different from factorisation) into a quantum code?

Well, least but not last, reading the libquant (freely inspired from quantum mechanics to compute risks) I can't help that notice that most so called smart boys with degrees in maths are totally clueless about quantum mechanic, because quantum mechanics is based on probability, and most human don't grasp it. So how can I trust less than able non -as smart and honest person- as traders to grasp QM?