Einstein found it creepy, referring to it as "spooky action," and while his reasons are likely different than yours, there's no doubt that the upcoming quantum computing revolution is likely to unsettle most.
In the span of a few short years, the tech that protects trade secrets, government files, your online banking, the stock market, shopping, and even Bitcoin will likely be rendered obsolete. This is a race where anything but first place is last.
Join Dr. James Walden and Dr. Kevin Kirby as they share a few qubits from across the galaxy, their thoughts, and a cup of coffee with host Mike Nitardy, as they untangle the topic of quantum computing.
Learn more about the College of Informatics at https://coi.nku.edu.
Einstein found it creepy, referring to it as "spooky action," and while his reasons are likely different than yours, there's no doubt that the upcoming quantum computing revolution is likely to unsettle most.
In the span of a few short years, the tech that protects trade secrets, government files, your online banking, the stock market, shopping, and even Bitcoin will likely be rendered obsolete. This is a race where anything but first place is last.
Join Dr. James Walden and Dr. Kevin Kirby as they share a few qubits from across the galaxy, their thoughts, and a cup of coffee with host Mike Nitardy, as they untangle the topic of quantum computing.
Learn more about the College of Informatics at https://coi.nku.edu.
Spukhafte Fernwirkungen. Did I say that? Spooky action at a distance?
James Walden:Yeah.
Kevin Kirby:Oh dear.
Mike Nitardy:What is that? What does that mean?
Kevin Kirby:It's one of the early bits of weirdness people noticed and quantum mechanics, something in particular disturbed Einstein. It's these very strange long distance correlations, where you can have a resource — Dr. Walden and I are sitting together and we, we share a couple qubits, say, for example, then we go, I go out to the Andromeda Galaxy, he goes to the other side of the Milky Way. And the statistical behavior of his qubit and my qubit are weirdly correlated in ways that are much different than if we split any other resource. There's sort of an over correlation. And it's almost like my qubit and his qubit were talking to each other faster than the speed of light. But they're not. And that is impossible. But there's correlations that make it seem like that is true statistically. So Einstein first used the term spukhafte Fernwirkungen. It became sort of a slogan associated with the weirdness of quantum mechanics.
James Walden:Yeah, Einstein didn't really like quantum mechanics. And so he did a lot of research trying to disprove it, and instead made tremendous contributions to the field by showing that these weird things actually are true.
Mike Nitardy:Well, hello, and welcome back to the Informatics Cafe. I'm your host, Mike Nitardy. And I'm so pleased and proud to have with me here in the cafe today, Dr. James Walden. He's a PhD in physics and he is the director of the Center for Information Security here at Northern Kentucky University. And also Dr. Kevin Kirby. He's the Dean of the College of Informatics here at Northern Kentucky University. And gentlemen, it's great to have you here in the Informatics Cafe with me, and we're going to be talking about the exciting field of quantum computing. So welcome. Thank you for being here today.
James Walden:Thank you, Mike.
Kevin Kirby:It's a pleasure to be here, Mike.
Mike Nitardy:Why don't you start off and give us your elevator speech — if there is such a thing — as to what is quantum computing?
Kevin Kirby:Well, I'll start with a value judgment. As a computer scientist, I think quantum computing is the most interesting thing to happen in computer science, since there was computer science.
Mike Nitardy:Wow!
James Walden:Quantum computing is computing using a device that maintains its internal state, basically, as an indeterminate state.
Mike Nitardy:Right.
Kevin Kirby:You have an undisturbed system, and you have computation going on. But as soon as you open the box to look at it, it sort of collapses.
Mike Nitardy:Okay.
James Walden:You can't tell anything about its internal state until you observe it.
Kevin Kirby:There's this interesting power that comes from quantum mechanics. Someone once said quantum mechanics is what happens when nobody's looking.
James Walden:Observations don't necessarily refer to a human looking at it. It is really, if any small particle hits the atoms involved in the quantum computation that counts as an observation.
Kevin Kirby:But the power you get from that, and the new understanding of information, we're so used to thinking in computer science of bits, ones and zeros ons and off, and..
Mike Nitardy:Exactly.
Kevin Kirby:... building computers out of those principles, to have a whole set of new principles with a little sort of, I don't know, almost supernatural mystery to it...
Mike Nitardy:Exactly!
Kevin Kirby:... is very compelling. The bottom line is there is the promise that we can do computations, amazingly, mind bogglingly faster than we could have imagined without quantum technology.
James Walden:They can't compute anything that our classical computer can't. But they can do certain tasks much, much faster.
Kevin Kirby:That's why it's attracting investors and researchers.
James Walden:Uh, maybe I should explain what a qubit is. So quantum computers compute using qubits — quantum bits — rather than our traditional bits. So a bit is just a zero or one, which can be represented in a lot of different ways physically. So normally, in your RAM it's represented by the charge. Then your memory chip's inside your computers, and if the charge is a certain level, it's a one if it's a different level, it's a zero. And it's either one or the other. So the charge can be slightly higher or lower. But basically, there's a threshold. If it's above the threshold is one, if it's below the threshold is zero. It is very clear cut. But qubits can be in a superposition of state. It has a certain chance to be a zero a certain chance to be a one. And as it interacts with other other qubits, those probabilities change. And it remains that way until you measure it. That's called collapsing the wave function and then you get either a definite zero or a one result. So you can read off the end result. But while you're doing the computation, the qubits are in this probabilistic state whether they're either a zero or one. The classical statement of that problem in quantum mechanics is Schrodinger cat, where you rather cruelly seal a cat in a box with a poison that's activated by a radioactive sample. Whether a radioactive sample emits a particle or not is a purely physical probabilistic thing. You can't say it's going to emit in one second, two seconds. Until you observe it you don't know when that happens. And so the cat in the box is either alive or dead. And you can't tell it without opening the box.
Kevin Kirby:And it's not just that the cat is alive or dead, right? It's, if you could flip a coin and say it's alive or dead. No, it's some, it's sort of sort of both at once, and you can detect the difference between a cat that's — flip a coin alive, or 50 percent alive 50 percent dead, you know, or in some way, both — you can actually pass them through a certain quantum gate that will give different results depending on whether it's in the spooky superimposed state or not. And that is just oh, my gosh, it makes you dizzy to think about.
Mike Nitardy:I was gonna say...
Kevin Kirby:It's why we're in the business, Mike.
Mike Nitardy:Exactly right. Exactly right.
James Walden:But the cool thing is that that intermediate state can perform computations extremely faster than classical computers with regular bits can do.
Kevin Kirby:My second favorite part about qubits is you can't
Mike Nitardy:That's amazing. copy them. How many times do you go through a day and hit copy and paste in your document? You can't copy sets of qubits, there's something called a no cloning theorem. Really? Wow! How do you do, how do you do computing without copying stuff? That's why it's hard. So does that mean that it's hard to replicate? I mean, so if if...
Kevin Kirby:A lot of a lot of our algorithms, you copy the contents of this variable to this variable.
Mike Nitardy:Right.
Kevin Kirby:You don't even think about it when you're writing code.
James Walden:Yeah.
Kevin Kirby:You can't actually do it, not reliably, in quantum computing. They're fundamental limits.
James Walden:Yeah, and it would also have impacts on debugging a program. So normally, when you're debugging, you're looking at the intermediate internal state and trying to figure out what went wrong. But of course, if you look at the internal state of a quantum computer, it collapses and you no longer have the internal state you wanted to observe. Because you because exactly that you looked at it. The computation stops at that point and can't be restarted.
Mike Nitardy:Wow. Wow. So essentially, we're at the cusp of something transformative that is exciting, because it's like, a new birth of computer science almost.
James Walden:It is. Yeah.
Kevin Kirby:Yeah. We're both saying yes to that. Sounds tidy but I think it's true.
Mike Nitardy:So I think we're we have to go back down to my level. [laughter] So Dr. Walden, why don't you go ahead and give us a little bit about your background.
James Walden:Okay, I got interested in quantum computing when I heard about it in graduate school, when Peter Shor was giving his talk tour at universities about Shor's algorithm, which provides a way on a quantum computer to factor an integer into its prime factors. So you can express any integer as a number of prime numbers multiplied together. It's very fast, of course, to get the product, you just multiply two numbers together. But to go backwards and take a large number and figure out what two prime numbers were multiplied to get it is very difficult. And before that, no one had an argument to why quantum computing could be better or faster than classical computing. But he invented the first algorithm for that. You might think that's an abstract mathematical problem to factor a number, it's like, who cares. But all of our E-commerce, software downloads and such are validated by digital signatures. And the algorithm for that depends on it being slow to factor a number. And so if quantum computing becomes feasible, we have to find a new system for securing everything on the internet.
Mike Nitardy:What about your role in in physics, in your background in physics, does that play anything?
James Walden:Quantum computing, certainly back in the 20th century, was really more of a physics problem than a computer science problem. There weren't any of the traditional tools, principles, and such that really feel more part of the computing field. That's changed since then. But still, there's, you sort of need a basic understanding of quantum mechanics, which most computer science students don't take... Right. No that makes sense....to understand how it works. There are efforts to do things like quantum programming languages, and such to hide the physics behind it. But right now in the field, you really have to have an understanding of quantum physics as well as computation.
Kevin Kirby:And actually, Dr. Walden should point out that not only does he have a PhD in particle physics from Carnegie Mellon, he actually worked for Intel. So he's done both the Q and the C in quantum computing.
Mike Nitardy:This is very cool.
Kevin Kirby:This is me as Dean.
Mike Nitardy:No no I know. I love it. No, it's great. It's great. So So how did how did you get interested in it? Other than being obviously the Dean of the College of Informatics, and it is the newest hottest thing since computer science got started?
Kevin Kirby:Yeah, well, my interest in physics is sort of strange. I had a really interesting ninth grade astronomy teacher in Detroit, he handed me a book called The Tao of Physics, you might remember it. And it had a spread in the middle of the book, just two images and one was a black and white image — I think it was the Upanishads written in Sanskrit, and the other one was a long, it was like a field lagrangian or something from particle physics. And there are two things that it... Well, certainly as a 14 year old I had no clue. But they were beautiful and mysterious. So I said I want to know what what those mean. And so I was attracted to physics because it was weird and mysterious and incomprehensible. I wasn't particularly good at it. I went on to get a PhD in computer science, but I was always interested in natural computation, biological computation. But I did manage to do a cognate in physics for my PhD, which means very, very narrow. And actually, I did specialize in quantum mechanics. So later, when quantum computing came around, it was just, oh, I love this stuff. I want to teach it. I want to hire people who can do it.
Mike Nitardy:So like I said, I usually don't talk about myself, but I'll give you a little bit of background. I'm a strong liberal arts kind of a guy, even though I got my degree in finance. So I like numbers. But I don't do a lot of computations. I'm a lawyer. My experience with computers is just sitting in front of them and having them do whatever I want them to do, or at least what I asked them to do or try to get them to do. To what extent will quantum computing change computing right now, as we know, it?
Kevin Kirby:Depends what you mean by right now. If you mean technology, very little. I mean, a lot of the literature compares quantum computing right now to the Wright Flyer in what was it 1903, the Wright Brothers plane, but it's almost like, yes, you see this sad little plane making a hop. But the potential is so big people are already developing flight reservation systems and modern airports with 40 gates, and so on. Because the promise is still there. The ideas about quantum algorithms date from the 90s, Dr. Walden was talking about a few classic ones. But they were very much pencil and paper, but people have started to build this stuff. It's a few to several years out, but progress is so fast. And if they manage to build things at a certain scale, yes, codes will be broken optimization problems will be solved super, super fast. Logistics and supply chain problems are a classic one, now. So within the horizon of a few years, it probably will be transformative.
Mike Nitardy:So does that mean that the way that my mobile phone works, the way that my laptop works, the way that our internet works today is that going to change if quantum computing lives up to you know, all of this promise that we're discussing?
James Walden:Yeah, you won't have a quantum computer on your desktop. The physics requirement of the super cooling, and such just won't work at either a price or physical scale that you would want one in your office. But there'll be more like, sort of an alternative type of supercomputer to solve really complex problems. So now we have all these high performance computing clusters and centers, probably there will be quantum computers added to that.
Mike Nitardy:What I'm thinking in my mind then is it's almost like a Back to the Future in the sense of, if you go through the history of computers, you go through these larger computers down to the smallest computers. And that's we're so proud to get there. But what it sounds like to me is that there's going to be these special computer rooms, again, that have these quantum computers, whereas the rest of us aren't going to have access to that. Is that accurate?
Kevin Kirby:Perhaps, but that's, that's normal now, right? I mean, yes, we do have a server room right here in Griffin Hall, in the College of Informatics which looks cool and Star Trek like, but a lot of our students are using, say, Amazon Web Services or other places where their actual computation is going on, for example, in our machine learning course. So in fact, right now, through Amazon Web Services, I can write some code here in Griffin Hall in Python and spin it up on a quantum computer somewhere.
Mike Nitardy:So you can potentially get access to quantum computers, just like you could through the cloud right now.
Kevin Kirby:Yep.
Mike Nitardy:To any okay.
Kevin Kirby:Right, it's basically another cloud service.
Mike Nitardy:Okay. So they're going to speak to each other, there's going to be a a way for, you know, old classic computers to understand what quantum computers are saying and doing.
Kevin Kirby:Yes, exactly. I mean, I think, I mean, you use... Well there are some specialized programming languages — what is it Q Sharp I think is Microsoft's quantum computing language — but they're also their libraries for very, very familiar programming languages that you can write your code in to develop quantum circuits. And then run them on quantum computers elsewhere made by different companies — both the big names like you know, IBM, Google, Microsoft, but also, some specialized companies are out there building quantum computers.
Mike Nitardy:But it sounds like the number one thing here is the speed. Is that really the promise?
James Walden:Yes.
Kevin Kirby:Yes.
Mike Nitardy:And without the speed, is there any benefit to it?
Kevin Kirby:I think there's an intellectual benefit. I mean, I think the notion of, of what it teaches us about what information means is very important.
Mike Nitardy:Nice.
Kevin Kirby:And it's fed back into the heart of physics. I mean, there's some... Quantum mechanics is a bunch of calculational rules, and a lot of people — I mean, it's hard for humans to understand what it's actually describing. And there's been sort of a reverse effect where thinking about information speaking as the College of Informatics, has helped make the foundations of quantum mechanics clearer. So I think there's always going to be intellectual stuff and head scratching stuff, even if we can't build fast machines, but we will build fast quantum computers.
Mike Nitardy:All the rage, it seems like now in the news, and the financial world is blockchain and crypto. And so what does quantum computing and the speed with unlocking everything do with that world?
James Walden:Right, quantum computing, assuming we can build a large enough one to do the computations, can completely Wow, that doesn't sound very good for the crypto world. break the security of blockchains that basically all cryptocurrencies are based on. You would be able to generate new blocks very fast. And basically, the way blockchain works is that the longest blockchain wins. So there's always this competition with multiple groups trying to add the next block to the blockchain because you get a reward in cryptocurrency for doing so. And that's how your transactions get added. These people miners, bundle them up, put them in a block and try to compute the correct block the fastest. Quantum computing would give you an unmatchable advantage in that. And if you can win that race, you can do things like unwind transactions, double spin your bitcoins, and so forth.
Kevin Kirby:It's disruptive.
Mike Nitardy:I was gonna say, talk about a disruptive technology.
Kevin Kirby:With a capital D.
Mike Nitardy:Let's bring all this you know, home a little bit to us here in the College of Informatics. What are we doing here in this area right now?
Kevin Kirby:Well, James, do we write a grant proposal to get a couple of D-Wave machines down the hall here in Griffin Hall.
James Walden:That could be fun.
Mike Nitardy:[laughter]
Kevin Kirby:Probably got.
Mike Nitardy:We just push.
Kevin Kirby:Teach a course. I think it needs to get into the curriculum in computer science. I mean, companies who aren't even tech companies are thinking about quantum computing. They're preparing for the day, where blockchain breaks, where everything that relies on crypto breaks, prepping for the quantum world. And our students need to do that with skill set. So one of the limiting factors in the growth of quantum technology is the skill sets from students. As a university one of the exciting things about quantum computing is you can do a lot of it with just a computer science background in some relatively elementary math, say linear algebra. Some of the code you can write with these toolkits, is accessible. So I'd love our students to come out with a with a taste of that. It's part of looking forward, it's part of what we do at NKU.
Mike Nitardy:Excellent.
James Walden:On the security side, we do teach students about post-quantum cryptography. So we don't really explain how quantum computation works in detail, but we give sort of a broad sense of what it is, and how it provides the speed ups. And this is leading to all this research and post quantum cryptography and causing people to use longer cryptographic keys. Current quantum computers are a long way from breaking modern key sizes. They would need to have around a million times as many qubits and the error correction facilities would at least need to be 100 times better than they currently are. But even with that the National Security Agency and National Insititue of Standards and Technology have issued standards for post-quantum key sizes. So people are already starting to adapt to make it take a longer time for quantum computing to catch up while people are developing these newer encryption algorithms that won't depend on problems that are easily rapidly solved by quantum computers.
Mike Nitardy:How far off are we from this being a reality for our everyday lives?
James Walden:I'd say we don't really know yet. I recently read the National Academy of Sciences report on quantum computing. And they were tasked to give a timeline and they basically said that they couldn't was their summary that there's a certain breakthroughs we need in things like quantum error correction, and just how to physically build a quantum computer.
Kevin Kirby:But of course, people are still doing proof of concepts now. A lot of companies are investing in that even at the small scale.
James Walden:There's there's currently a wide variety of approaches to physically build them and we don't really know which of those will be successful scaling up if any, or whether we'll need to find new physical principles to build them on. So we don't really know yet.
Kevin Kirby:So we had a group of visitors here in the College of Informatics from the Fidelity Center for Advanced Technology in Boston. And I stumbled across their work where they were using quantum computation to simulate securities. They were doing an optimization problem, what's the right mix of pretend stocks to optimize returns in a portfolio. It's got zillions of variables, and you want to do something called annealing to find a solution. And that's one thing that quantum computers may be very good at long before they can sort of break blockchain and crypto. So you see, companies starting to get their feet wet, and they're starting to invest in that and train their people in that.
Mike Nitardy:That's obviously going to be a very disruptive technology. How do you know when you're getting a return? You know, if you're starting to invest in it, is it just for the expectation that you might, you know, land something it at some point? Or is there an actual expectation of return and getting some money to make off of it within the next 10 to 15 years?
Kevin Kirby:It's down the road, but people invest with long horizons occasionally.
Mike Nitardy:No, that's true. That's exactly right. That's exactly right.
James Walden:And certainly, the National Security Agency is heavily investing as are the intelligence agencies in China and Russia and other major countries because they don't want to be the last person who's able to break all encryption.
Mike Nitardy:No doubt, I would imagine where they are on the progress. And it's probably very guarded,
James Walden:I think, right now that ability is the main thing that's attracting funding to it. But because I believe we will have post quantum cryptography in the next decade or two, quantum computing needs to find other solutions. There's several types of quantum computers and the big division is between digital and analog.
Mike Nitardy:Okay.
James Walden:And that used to be true with classical computers. I don't know if anybody still has an analog, traditional computer anymore. But in the mid 20th century, they could solve certain physics problems, differential equations and such, faster than digital computers of the time. But digital computing sort of ran away, growing rapidly in performance following Moore's law. And so analog classical computers were sort of dropped. But with quantum, we're not really at that point yet. And so mainly, I've been talking about the digital ones because that's where you get the focus on attacking encryption and blockchain and the like. Whereas Kevin was talking about simulated annealing, which is something that a analog quantum computer can do.
Mike Nitardy:I'm so overwhelmed by the brainpower that you guys bring to the table here and just to talk about this. I could sit here the entire day but I know that we all have other things to do. In the next five years, any major breakthroughs, anything that would change the playing field for a major company to come out and say, we've done X, and this has changed everything.
James Walden:Google did make such a claim in 2018, of so called quantum supremacy, which means that you've performed a computation on a quantum computer faster than any classical computer could do. But that claim is still being disputed.
Kevin Kirby:Quantum primacy or quantum advantage, trying to find a better noun than supremacy right now for that, but it will always be contested.
James Walden:I read an article this month about a group at a high performance computing center. He said like, well, when we do with using this algorithm, our result is faster than Google's. And so...
Kevin Kirby:Right? I mean, so far, these examples of problems, where a quantum computer seems to be, you know, millions of times faster than the classical computer are basically sort of almost like simulating physics problems. So it's like, you're
almost cheating:Physics doing physics.
Mike Nitardy:[laughter]
Kevin Kirby:Of course, somewhat faster. But yet, then you have people come back with conventional computers and try and beat it.
James Walden:I suspect we'll see more claims of quantum supremacy and more disputes about it. I'm not sure if we'll get a clear cut answer in the next five years or not.
Kevin Kirby:I think there's always going to be this empirical race and, and that's, that's going to be fun to watch...
Mike Nitardy:Right, right.
Kevin Kirby:..actually benchmarking real quantum computers on more and more realistic problems. So those are the headlines we're going to read over the few years, they're going to be dramatic.
Mike Nitardy:The brain power in the cafe today is in overdrive. Thank you both so much for joining us today. I've just been so humbled just sitting here with you both talking about this awesome topic, and I hope that our listeners have benefited from it as well. I'm sure that they have. Thank you both.
Kevin Kirby:Thank you, Mike. It's so fun.
James Walden:Thank you Mike.
Chris Brewer:Informatics Cafe is presented by Informatics+, the outreach arm of Northern Kentucky University's College of Informatics. Hosted by Mike Nitardy. Produced and edited by Chris Brewer. Music and recording by Aaron Zlatkin. Recorded at the Informatics Audio Studio in Griffin Hall.