Quantum Bayesian Networks

November 24, 2013

Quantum Computers and Canadian Wimpiness

Filed under: Uncategorized — rrtucci @ 8:23 pm

“The future belongs to Canada”, according to MacLean’s’ Mark Steyn

quote from TV program “The Simpsons”: (said by Grampa Simpson) “Typical Canadian wimpiness. That’s why you have snowballs and we have the H bomb!”

MacLean’s, a famous right-leaning, franco-phobe Canadian magazine, has just come out with a news article containing their “Power List” of the 10 most powerful Canadians. Mike Lazaridis made the list (by a hair, as number 10). MacLean’s article delivers a well written description of Lazaridis’s efforts in the quantum arena. Well written but not very critical. Me being such a nag, let me provide some of the criticism that is lacking from their article. In particular, let me point out some typical Canadian wimpiness aspects of Lazaridis’ quantum computing efforts.

Here is the article:

How did Mike Lazaridis make our Power List?
by Paul Wells (MacLean’s, Nov.21, 2013)

Excerpts in boldface. My commentary in non-bold

In 2002, Lazaridis launched his project’s experimental phase, the Institute for Quantum Computing (IQC). Spilling across three buildings including the year-old, state-of-the-art 280,000-sq.-foot Quantum Nano Centre, IQC is home to 200 scientists trying to turn theory into working prototypes for a new technology.

When Lazaridis hired a soft-spoken, Quebec-born student of Stephen Hawking’s named Raymond Laflamme to be the director of the Institute for Quantum Computing in 2002, Laflamme had built a five-qubit “computer,” an incredibly finicky tinker toy. Laflamme is up to a dozen qubits now; he hopes to have machines with 50 to 100 qubits in the next five years. To get this far, Lazaridis has paid $200 million of his own money, an amount roughly matched by a series of investments over 13 years by the Ontario and federal governments.

Laflamme’s 12 qubits are NMR qubits. NMR quantum computers were known to be non-scalable since they were first proposed about 17 years ago. For many years, Laflamme has clung doggedly to them and managed to push them as far as they can go. A great waste of time, since NMR QCs are a non-scalable dead end street.

According to Wikipedia, Laflamme’s original training is as a pure theoretical physicist in quantum gravity and cosmology. Leading a team that invents a practical QC will require an experimentalist of the highest caliber, one guy in a million. Most of the time, great experimentalists and great theorists in physics have very little in common: they are a different species, with different upbringings, interests and talents. There have been some exceptions to this rule of thumb, like Enrico Fermi, but they are very rare. Is Laflamme destined to become Canada’s Fermi? Highly unlikely, unless he is a very late bloomer.There are ZERO U.S. Patents (ref:below) that list him as an inventor even though he is 53 years old (ref: Wikipedia). If he were going to show promise in the experimental arena, he would have done so at a much earlier age. Really great experimentalists are already doing jaw dropping experiments while they are still in high school. If you were trying to assemble a winning hockey team, would you want the team’s coach to be a 50 year old, klutzy, book worm who first developed an interest in sports a few days ago?

Laflamme showed me sheets of factory-made diamond with precisely inserted impurities, individual atoms poked into the carbon lattice with microscopic accuracy. The resulting material has quantum properties. He showed me a sensor made of silicon and etched aluminum. An IQC researcher, Adrian Lupascu, had an insight: If quantum states are incredibly delicate and prone to break down, their very fragility can make them useful for sensors. “These are the best magnetometers there are, the size of microns . . . an order of magnitude more sensitive than anything else.” They’ll have applications in medical imaging, manufacturing and microscopy.

In other words, 11 years after they started in 2002, the best they have to show for a 200 million dollar investment (plus 200 matching grant from the government) is a magnetic sensor. They are a million miles away from constructing a quantum computer. They aren’t even focusing on that.

Laflamme has been recruiting world-class scientists at an accelerating pace: materials scientists, engineers, theorists. His biggest catch by far is David Cory, a nuclear engineer the University of Waterloo poached from the Massachusetts Institute of Technology in 2010 with a $10-million federal grant. Nobody in the world is better at understanding the odd language of quantum particles or in spotting their real-world applications. He came to Canada with three 18-wheelers full of lab equipment. That posed the sort of challenge Lazaridis likes to solve.

Unfortunately, David Cory ‘s specialty is mostly NMR. Cory has worked on quantum computing for at least 17 years (I located an NMR QC paper of his from 1996). The types of QC advances that he has made in those 17 years are almost exclusively in NMR, a non-scalable, dead end street for quantum computers from day one. A patent search shows only 1 U.S. patent (ref. below) with him as inventor, not too impressive for a fairly old experimentalist. (I couldn’t find his age on the Internet but I believe he is 50-60 years old).

Cory has an unusually large team of graduate students and post-doctoral assistants from a wide variety of disciplines. The entire team meets every morning. Nobody is allowed to work on individual projects. They have found applications for their work in medical imaging and in deep-sea oil exploration. “We’ve started exploring with the Princess Margaret Cancer Centre” in Toronto, Cory said. “Just an open-ended discussion. How can quantum sensors make a difference? And there’ll be many ways. It’ll build on old work we did at the Dana-Farber cancer centre in Boston, using quantum sensors to set the edges of soft-tissue sarcomas.”

Just what I suspected. Cory wants everyone in his team to follow his vision (which seems to comprise mainly what is familiar to him, i.e., NMR related stuff). They are working almost exclusively on quantum sensors, which is quite different from working to build a QC.

In September, IQC named its two latest recruits. Amir Yacoby comes from Harvard University and is a world leader in developing these exotic materials that can hang onto quantum characteristics. Steve MacLean is an astronaut: he served on the International Space Station in 2006 and was the president of the Canadian Space Agency from 2008-13. But MacLean is also a physicist who was setting up an elaborate experiment in his garage—he won’t discuss details—when he bumped into Lazaridis a few years ago at Cape Canaveral before a space shuttle launch. “We ended up talking for two hours about everything and nothing.” Soon MacLean had offices at Perimeter and at IQC. He also serves on the advisory board of the new venture-capital firm.

According to Wikipedia, Steve MacLean got a Ph.D. in laser physics in 1983 (at the age of 29, fairly late for typical PhDs), but most of his career has been as an administrator, gymnastic athlete, and astronaut.There are ZERO U.S. Patents (ref:below) that list him as an inventor even though he is 58 years old (ref:Wikipedia).

As for Amir Yacoby, he seems to have been an academic most of his life. I found only 3 U.S. Patents (ref. below) listing him as an inventor, again not too impressive for a fairly old experimentalist (I couldn’t find his age on the Internet but he looks from his photos to be 50-60).

Quite frankly, judging from their WIMPY patent record, I do NOT expect either Laflamme, or Cory, or MacLean or Yacoby to be the first scientists to build a practical gate model QC.

What are they doing? MacLean won’t say.

“You know what I’m doing by doing that?” MacLean said in an attempt to explain his silence. “I’m protecting Canada in a way. I mean, this is a community effort that we’re doing here and we have certain leads in certain areas and I just would like you to focus on the fact that it’s unique.”
Only Lazaridis can talk about what goes on at QVI. He doesn’t say much either. But he reveals that the fund has made its first investment. “Actually, we only went public after we did our first investment.”
What’s the investment?
“I can’t tell you that.”
What’s the scale of the investment?
“It’s big.”
What field is it in? Is it in sensors? Cryptography?
“Those were already announced before we made our investment. Our investment is different.”
Will it lead to products on the market? Yes, Lazaridis said, perhaps in two or three years. Products for consumers? “I can’t really tell you.”
He finally said that the entrepreneur QVI just funded is “a guy from around here” and that when he wrote an equation on a blackboard, “I said, ‘We have to invest in this.’ This is not just an improvement, this is not just a new way of doing a classical thing that we’re all used to. This is something new.”

Well, I think that such excessive coyness from Lazaridis is counterproductive and self-defeating. That is what patents are for, to allow partial disclosure without losing all rights to an invention. Even Google started with a patent. Do Lazaridis’ people have any patents at all to back up all their hot air? Just by saying that they are working on a magnetic sensor or a photonic QC or whatever, (and then pointing to some patents) would not give the store away. It would actually garner free publicity and serve to enlist new investors and scientific talent. If I were an investor, I would be very leery and put-off by such excessive coyness on the part of a startup, because it makes them indistinguishable from scam artists, or deluded dreamers.

Furthermore, I would argue that in their case such coyness is not protecting them from any competitors because they are operating in an essentially competitor free environment—I mean, how many other companies are trying to build a QC? (other than D-Wave, and Lazaridis seems totally uninterested in adiabatic quantum computers like D-Wave’s) Of course, I am assuming that Lazaridis and his coterie mean it when they say that they are trying to build a QC. The truth is they don’t seem to know themselves what they want. They claim that they are determined to build a QC, but they claim this only 3 out every 7 days of the week. Canadian wimps!

Patent Searches:

Wikipedia Bios

November 19, 2013

Bayesian Networks: a Programming Language For Sharpening Fuzzy Probabilistic Thinking

Filed under: Uncategorized — rrtucci @ 8:20 pm

Lately, I’ve been busy writing 3 papers (for Operation Lisbeth). In the process, I’ve had an “epiphany” experience which I’ve had several times before and which I’d like to describe to you. It’s an experience related to Bayesian networks (B nets). Often, when I have questions of the type: quantum mechanics predicts such and such doesn’t it? Or what does quantum mechanics predict for such and such?, I find that B nets are an excellent “programming language” to answer such questions in a very conclusive and satisfying way, at least for my taste.

Everyone that has ever programmed a computer has had this experience: You start with an algorithm that you think you understand, but after programming it on a computer you end up understanding it much better than before. The process of programming it makes you aware of a lot of subtleties about it that you didn’t realize before.

I find the same is true when I use B nets. I start with a crazy probabilistic algorithm that I think I understand. Then I “program” it in the language of B nets, and all of a sudden I realize how poorly I originally understood it. Often my original algorithm was slightly wrong or incomplete, and the language of B nets comes to the rescue, allowing me to fix the parts that didn’t work at first.

Note that by “programming” my crazy algorithm in the language of B nets, I don’t necessarily mean writing an actual computer program for it. I mean simply stating it in the precise language of B nets. A computer program is not necessary, even though going that extra mile might produce extra satisfaction. Computer software for doing both classical and quantum B nets does already exist, and I believe it will continue to improve in the future. I even believe that quantum computers will revitalize the field of B nets, by allowing us to perform B net calculations much faster, both for classical and quantum B nets.

I have found that this “programming” with B nets strategy works for BOTH classical and quantum mechanical probabilistic algorithms.

For classical algorithms, I program in the language of classical B nets pioneered by Judea Pearl and others. (For an example of this strategy, see for example this paper of mine, in which I review classical Shannon information theory, stated in the language of classical B nets.)

For quantum algorithms, what I like to do is to FIRST find the classical version of the algorithm, and program that in terms of classical B nets. SECOND, I generalize the classical algorithm to quantum mechanics, and program that in the language of quantum B nets (quantum B nets are discussed in this paper). As an example of this TWO STEP STRATEGY, see, for instance, this paper of mine about Maxwell’s Demon.)

Related Blog Posts

November 6, 2013

Google Planning to do Teleportation

Filed under: Uncategorized — rrtucci @ 6:01 pm

Secret Message: Operation Lisbeth is going very well

Recently, some press articles have come out claiming that a mod of Minecraft called qCraft, produced by Google, will cause millions of kids to get interested in quantum science and become the quantum scientists of the future. Maybe, or maybe not. Maybe they’ll get hooked on video games and end up as operators of military drone planes.

One of the things that qCraft can “do” is quantum teleportation. This has led some of those future scientist kids to think that the reason Google currently has a barge floating in the San Francisco harbor is that it is planning to teleport it, in broad daylight, into a different universe, using Hartmut Neven’s multiverse theory and D-Wave’s quantum computer. Google has been practicing how to do this with their qCraft quantum simulator. According to qCraft, it should work.

No way kids. D-Wave’s QC is incapable of teleportation, because the degree to which it can entangle things is not sufficient to achieve it. Teleportation can only be achieved by gate model quantum computers, not by adiabatic quantum computers… at least perfect teleportation. At most, adiabatic QCs like D-Wave’s might be able to achieve an imperfect teleportation. Let me illustrate this with an animated gif of the inferior teleportation product sold by Google. If you want perfect teleportation, you can’t have it kid, because Google refuses to work on gate model QCs. Google could invest in both QC types and race them against each other but it hasn’t. Google is no Bell Labs, it’s not even an IBM, it’s more like Ron Popeil‘s Ronco-matics lab.

Blog at WordPress.com.

%d bloggers like this: