Quantum Bayesian Networks

May 17, 2015

Quantum computer devices, Report Card

Filed under: Uncategorized — rrtucci @ 8:19 pm

Quantum computing didn’t really take off in any serious way until Peter Shor published his factorization algorithm in 1996. Hence, we are quickly approaching the 20th anniversary of quantum computing’s inauguration or sendoff. This month is also the end of the school term and the beginning of Summer vacations in the US. Hence, I thought it would be fitting to give quantum computer “makers” a 20 year report card.

There are two fat students taking the course “Quantum Computer Devices 101”. These two students are called Universities and Industry. Here are the grades I give them:

Student Name: Universities
Course: Quantum Computing Devices 101
Grade: F

  • MIT: Boson Sampling
  • Caltech: QCraft
  • IQC at Univ. of Waterloo, Canada: NMR quantum computer and quantum cryptography
  • Joint Quantum Institute – Univ. of Maryland
    student refused to submit test paper. Claimed his answers were Top Secret.

Student Name: Industry
Course: Quantum computing Devices 101
Grade: B+

  • D Wave: Steady progression of quantum annealers based on superconductive Josephson devices
  • Google: Has used D-Wave devices in past. Now building both a quantum annealer and a gate model QC of their own design, both based on superconductive Josephson devices.
  • IBM: gate model QC based on superconductive Josephson devices
  • Microsoft: Anyonic quantum computer
  • Intel: the biggest underachiever of the Industry class, a disgrace to his parents who are working so hard to pay his tuition. Although very talented, has done no QC device research. If only he stopped goofing off.

Grader’s Comments:

The pattern is very clear. Industry now rules the QC devices field. After 20 years of “trying”, universities have achieved very little in this field except for Martinis’ team at the Univ. of Calif. at Santa Barbara, but those guys now work for Google. Given the choice between continuing to work in the university system, or joining industry, Martinis ran as fast as he could to the industry camp. This speaks volumes about the American university research system.

  • Concerning MIT, the grader was being kind. In truth, MIT has not built a Boson Sampler themselves. They outsourced it to other universities outside the USA, either because their experimentalists are too inept to do it themselves, or too disinterested in the subject. Disinterest is a real possibility, as boson sampling has zero practical applications. It’s sole purpose is to answer an arcane, esoteric mathematical question.

    MIT did participate in building an NMR quantum computer (David Cory and Isaac Chuang), but that was more than 15 years ago. This grader has never understood why NMR quantum computers were built in the first place. Such computers were known to be non-scalable beyond 10 qubits from the very beginning. So they were known to be a dead end street from day one. Furthermore, papers were written showing that NMR “quantum computers” are so noisy that they are indistinguishable from a classical computer. Nevertheless, university academics hailed NMR “quantum computers” as a marvel of scientific ingenuity. Isaac Chuang was acclaimed as an experimental genius for building them, and he was made a tenured professor at MIT. Since then, Chuang has not produced any QC devices whatsoever of any merit. Instead, Chuang has dedicated himself to writing dishonest scientific papers. Dishonest scientific behavior is apparently condoned and even encouraged at MIT. He is a perfect fit for their staff.

  • Concerning Caltech, the grader was being kind here too. Caltech has been “trying” to build a QC since Shor invented his algorithm in 1996. After 20 years of trying, and burning through tens of millions of dollars, Caltech has built no QC devices except for QCraft, a “mod” (extension) of the computer game MineCraft. Actually, they didn’t even write the QCraft software themselves. Some teenager outside of Caltech wrote it for them. QCraft is very loosely based on quantum computing. It’s purpose is allegedly to teach quantum mechanics to children, but it hardly does that. It teaches a child as much quantum mechanics as a pinball game with the theme of soccer teaches a child to play the sport of soccer. No thanks. I prefer taking my child to a soccer field to play and making him/her join a soccer team.
  • Concerning IQC at Waterloo, Canada, the grader was being kind here too. IQC was founded in 2002. Since then, it has spent something like 200 million dollars trying to build a QC but has yet to make any significant advances in building one. They build great buildings though. 

    IQC is headed by Raymond Laflamme (its president for life?). Laflamme started his QC career more than 15 years ago at the Los Alamos National Lab working on NMR quantum computers, and he is still working on them to this day, even though they were known to be a dead end street from day one. ( Here is Laflamme’s latest paper, dated 2015, on NMR quantum computers). IQC is also devoting a lot of its energy to another dead end field, quantum cryptography.

  • NIST at Colorado has made some notable advances in QC devices, but NIST is a government lab, so it is neither Universities nor Industry. Furthermore, NIST’s ion trap devices have by now been surpassed by other designs based on Josephson junctions, etc.
  • The Australians have also made some progress in building a Bruce Kane type quantum computer. A parallel American project headed by B. Kane at Spook University (a.k.a. the Univ. of Maryland) seems to be going on, but is unfortunately classified (thanks to IARPA and NSA). According to documents leaked by Ed Snowden, Spook University has received from the NSA and other federal agencies in the neighborhood of 100 million dollars to build a QC. What did they do with it? Your guess is as good as mine. Perhaps mostly just build some mammoth buildings, something universities love to do (instead of paying their custodians and adjunct teachers a living wage.)

May 4, 2015

IBM vs. Google—Another Gun Fight at the OK Corral

Filed under: Uncategorized — rrtucci @ 1:32 am

IBM says: “Come out of there Google. Your ass is grass”

My friend Sol Warda prompted me in the comments to speak about a recent flood of news reports about a paper IBM published in the April 29th issue of Nature magazine. In the paper, IBM discusses their recent advances in error detection for their gate model quantum computer architecture based on Josephson junctions. Here is one of many such news reports.

I responded to Sol in the comments, but I am hereby promoting the response to a blog post. Here is what I said.

… I think it’s great news for QC software companies like Artiste-qb.net. Some of the quotes I’ve read from IBM spokesmen say something like “IBM will be the first to build a real quantum computer”. Now those are fighting words. Sounds to me like they are dissing annealers and challenging Google to a duel. It’s interesting to compare IBM’s gate model device to Martinis’ before he was convinced by Google to build an annealer. I think IBM may force Google to ask Martinis to build both a gate model device and an annealer at the same time. (They are sort of complementary anyways.) Big G sure has enough dough to do it, much more than Big Blue.

May 1, 2015

How Tara the Elephant and Bella the Dog Saved Mankind

Filed under: Uncategorized — rrtucci @ 7:01 pm

I try to keep this blog focused on quantum computation and quantum Bayesian networks. This blog post has nothing to do with either, so you better skip it if that is what you come here to read. The blog post is basically an overly sentimental children’s tale or an amateurish Greek-like myth about animals, written by me to avoid doing real work.

Blog at WordPress.com.

%d bloggers like this: