Recent comments from SciRate

Thanks should go to Noon Silk, who sent a patch to add MathJax support.

@Noon Silk. I do not know what to say about heuristic 3, but, in relation to your first question, let's assume that all heuristics are valid and apply theorem 2 to solve the Discrete Logarithm over Z_q*, where q is prime. As far as I understood (someone please correct me if I am wrong) the algorithm

...(continued)

"the entangled states in the GK theorem include the GHZ state and the measurements that you need in order to violate Bell inequalities for these states"

If this is true, the GHZ state and all local measurements' projectors needed for Bell inequality violation must be in the same MUB polytope, con

...(continued)

Physicists may be also interested in a realistic proposal of DFS computation with solid-state qubits: Phys. Lett. A 374, 3285 (2010), arXiv:0903.1056 [quant-ph]

This is almost identical to my previous work:

http://arxiv.org/abs/0904.1969
http://arxiv.org/abs/0906.4133
http://arxiv.org/abs/0909.2432

This Harvard-MIT Lincoln Lab paper claims that quant-ph/0303039 by Bullock and Markov was the first paper to give an exact decomposition of diagonal unitary matrices into elementary gates. That’s false. This Bullock Markov paper repeats the exact decomposition of diagonal unitary gates which was qui

...(continued)

The paper says
"To implement an n-qubit diagonal unitary exactly on a
quantum computer generally requires 2^{n+1} − 3 one- and
two-qubit gates [7]. However, one is usually interested
in circuits that approximate the unitary to within some
error tolerance, \epsilon." [7]=Bullock-Markov

NOW yo

...(continued)

Hi Aram,
Could you please tell me, what is the most efficient way known to project out the totally symmetric part from an arbitrary n qudit pure state?

Sorry, I now realize it's a dumb question. No need to answer it.

Hi Courtney and Steve,

I wouldn't say that it is a linear Bravyi-Haah, but I can see the similarity. Bravyi-Haah clusters all neighbours within a length scale, as Steve says, whereas I just do a pairing. I think that will result in quite different behaviour. The nice crossover point for the thres

...(continued)

Hey James,
It seems to me that this decoder is very similar to the Bravyi-Haah decoder but with a linear schedule of length scale increases instead of the exponential schedule as described by BH. Can I think of your decoder like this or is it an inaccurate comparison?

Thanks Oded, I will pass it on to my co-authors. The combination of the two results seems to give quite a strong argument that DQC1 is truly intermediate.

No.

I have not gone through all the details, but it seems that some of the results are already antecipated in the paper below, where they show
that states with discord created locally can not be used for entanglement distribution.

http://link.aps.org/doi/10.1103/PhysRevA.87.032340

WOW... these journals seem really suspect. The director of the publishing company, Prof. Mahmoud Abdel-Aty, has attracted some criticism on this blog (see the comments section): http://elnaschiewatch.blogspot.de/2010/07/somethings-going-on-at-amis.html

There is certainly some originality in the research of Abdel-Aty. For instance, he seems to be the first person to have gone realised that the concept of entanglement sudden death can be inverted to study entanglement sudden *birth*.

Also, seems that Barry Sanders has had some interaction with this Abdel-Aty before. According to:
http://elnaschiewatch.blogspot.de/2010/07/somethings-going-on-at-amis.html
Barry used to be managing editor of the AMIS journal, but is no involved in the journal!

From Anthony's search, I don't see any RMPs on fault tolerant quantum computing (excluding the review on TQC), and I can only see a passing mention of quantum algorithms. Perhaps I am biased, but I'd say these are cornerstones of quantum information theory.

This is an interesting addition to the repertoire of approaches to fault tolerant quantum computing, and not needing state distillation is very cool! I'm curious how this approach compares with using the [[15,1,3]] code with transversal pi/8 and transversal CNOT, and then implementing the Hadamard

...(continued)

This is a great sidestep around the Eastin-Knill theorem.

A nicer way to imbed the pure states into a $d^2$ dimensional space is to shift the origin to the identity (total mixed state). This makes the space a vector space rather than an affine space and two vectors are orthogonal if and only if the corresponding pure states have overlap $1/d$. Repeating th

...(continued)

We liked this paper so much that we made a short video introducing the main ideas behind it. :)

Naturally, this is just our personal take on the paper. As so, we alone are to blame for mistakes, misinterpretations and weird camera angles.

http://youtu.be/gtcPp7FY0gU

This is just amazing.

Since combinatorics isn't the most popular section on SciRate, here's a blog post about this by Gil Kalai:

http://gilkalai.wordpress.com/2014/01/16/amazing-peter-keevash-constructed-general-steiner-systems-and-designs/

Wilson said at today's seminar at Caltech that Kee

...(continued)

Is this saying that making things that have classical (i.e. computational basis) groud states more classical by using a repeitition code makes it easier for a "quantum" annealer to find this classical ground state?

Very nice paper, which includes an exhaustive optimization of possible protocols allowing for different distillation schemes at different levels of concatenation.

For their numerical calculations, the authors have restricted their attention to the case where the input error on the raw magic state

...(continued)

Thanks for your comments Steve. We were also pleased with how well the compressed sensing approach worked, particularly with its seemingly inherent robustness to errors.

With regards to epsilon, it is actually modeled as a time-dependent parameter with t^2 dependence (epsilon = constant*t^2). T

...(continued)

This might be ground breaking on quantum foundations.

Comments and related references are welcome to pfrimer.physics@gmail.com

I'd like to see the list of those 44 RMP articles since 2000 relating to quantum information, to assess how much ground they cover. (I could reconstruct it, but that's too much work.)

We would like to respond to the previous comment by rrtucci, and to clarify any misunderstanding of our work.

We do not claim that quant-ph/0303039 by Bullock and Markov was the first paper to give an exact decomposition of diagonal unitary matrices into elementary gates. To the best of our knowl

...(continued)

Searching the RMP website with "quantum information" in the full text gives around 50 papers:

http://publish.aps.org/search?c[][operator]=AND&c[][field]=author&c[][value]=&c[][operator]=AND&c[][field]=abstitle&c[][value]=quantum&c[][operator]=AND&c[][field]=fulltext&c[][value]=&c[][operator]=AND&

...(continued)

There was a discussion about a previous paper of Joux (with a weaker result) on this blog post:
https://rjlipton.wordpress.com/2013/05/06/a-most-perplexing-mystery

"Of course, ghosts do not exist..."
seems like a rather strange assumption in order to define "ghost world".

:-)

Funny-looking zebra in Fig 1(a)

And here is a question on cstheory.SE about this paper: http://cstheory.stackexchange.com/q/18134/1542

Nice paper indeed.

@Min-Hsiu Hsieh: for a more fair comparison, you should normalize by the total number of scirate users. Anyway, it's great to see that this number is growing quickly!

I thought this was long overdue. Thanks!

I'm confused. Presumably they are not claiming they have a counterexample to Bell's theorem. If not that, what does it mean to simulate Bell violations without quantum resources (assuming a Bell-local model)?

Let's go with the probabilistic version of the CHSH inequality,

(1/4)(p(a_0=b_0)+p(a_0=b_1)+p(a_1=b_0)+p(a_1!=b_1))<=(3/4).

This probability will clearly not exceed the 3/4 bound regardless of what values you assign to a_0,b_0,a_1,b_1.

Indeed, if you choose the variables to take values in {

...(continued)

They cite http://arxiv.org/abs/0908.3023 but I don't see how they address its criticism of the idea of making computational use of nonlinearity.

In particular, they respond to the paper by saying "In particular we
take the common view that if pure states are deterministically prepared and insert

...(continued)

Great title! I'm not convinced it's simpler, but it's nice to see an unfamiliar take on a familiar topic.

oh man...

There is something funny about that journal (and also many others). The landing page for the journal is here:
http://naturalspublishing.com/show.asp?JorID=16&pgid=0

It takes some clicking to get to the actual article(s) published in the journal. Much more prominent are the instructions to auth

...(continued)

The original quantum Pagerank paper used adiabatic evolution, and this one uses the Szegedy walk. I wonder how the methods compare.

Another thing I'd like someone to figure out at some point is whether these can be done using resources scaling like poly log(# vertices) for power law graphs. T

...(continued)

Han-Hsuan, I think Dave is using the term "Bell inequality" to refer more generally to entanglement witnesses that are can be constructed from correlated local measurements.

In the multipartite setting, these witnesses distinguish the entangled state from any tripartite separable state. Here's t

...(continued)

I'll answer anyway. :) This was written about in:
http://arxiv.org/abs/quant-ph/9604028

Their simulation produces outputs that are not +1/-1 valued.

Here is a toy version of their argument.
In normal CHSH, Alice chooses a_0, a_1 and Bob chooses b_0, b_1, with the goal of maximizing
a_0 b_0 + a_0 b_1 + a_1 b_0 - a_1 b_1.

Normally we have the constraint |a_i|, |b_i| <= 1.
Then t

...(continued)

I think that's basically right.

To be clear, they didn't do the "multiply by sqrt(2)" thing I said, but rather replaced +-1-valued observables with complex numbers. Still it does seem to miss the whole point of Bell's Theorem.

It's nice that there's a discussion of why discord is an important thing to look at. However, all of the examples are cases where the difference between zero and nonzero discord is interesting, and not examples of where *quantifying* discord (given that it is nonzero) tells us something useful.

This paper reads as though Dyakonov stopped reading the literature after the original FTQC paper by Aharonov and Ben-Or. In discussions a few months ago, I mentioned the following papers that address his criticisms:

http://arxiv.org/abs/quant-ph/0504218
http://arxiv.org/abs/1207.6131
http://ar

...(continued)