In this interview, Tonya Hall talks to IBM's Talia Gershon about the differences between quantum computers and traditional computers.
Tonya Hall interviews executives for our sister site ZDNet, and we're running a selection of some of her most viewed videos. The following is an edited transcript of her conversation with Dr. Talia Gershon, a senior manager of emerging technology experiences at IBM Research. (The interview was first published on ZDNet in July 2018.) To watch more of her videos, check out The Tonya Hall Show on ZDNet's YouTube channel.
Tonya Hall: What do you do in your role as senior manager of emerging technology experiences at IBM Research?
Talia Gershon: Well, my team is responsible for making quantum computing available in the cloud. That includes the open public devices we have on the IBM queue experience, as well as the cloud-based devices we have for our clients in the IBM Q network.
Tonya Hall: Give us a brief introduction to quantum computers. I mean, how do they operate, and what makes them different from traditional, or, I've heard you say, classical computers?
Talia Gershon: Quantum computers are a kind of computer, and just like any computer, they use physical operations performed on specific devices for computation. Our classical computers, they have bit-wise operations, which are communicated through voltage pulses to individual bits, and basically that's how you operate bit-wise on classical devices. But with quantum computing, we actually use different physical phenomena for computation. We'll use certain properties in quantum mechanics to process information in qubits or quantum bits.
We'll actually control the state of a device where that state is actually a quantum state. Instead of just a one or a zero, or an up or a down, kind of a spin up or down magnetic polarization like we have in classical devices, we'll use particle spin or other quantum phenomena, and we'll actually control properties like superposition and entanglement.
SEE: Quantum computing: An insider's guide (free PDF) (TechRepublic)
Tonya Hall: So what is a qubit? How do you use it to perform basic arithmetic? How do you control error rates?
Talia Gershon: So qubits are quantum bits, so they're a carrier of quantum information. We control them using microwave pulses in our case. The kinds of qubits that we use in our systems are super-conducting qubits. But you could use anything as a qubit, where you could control the quantum properties of that object. So if you trap an atom or an ion on a surface and you can control the properties of that atom or ion, you could use that as a qubit. But in our systems, we'll use superconducting qubits, and we control them using microwave pulses.
Tonya Hall: So what kind of environmental factors--like temperature or external radiation--affect quantum computers?
Talia Gershon: Quantum information is very fragile. Anything, like stray heat or radiation can actually influence the quantum state. We really need to shield our devices from all kinds of stray radiation or any external disturbances if we want to be able to preserve the information. We want to be able to preserve coherence times, for example, or other relaxation times or other properties for long enough to do computation. So we make it very, very cold. These systems live at the bottom of a dilution refrigerator at around 10 or 15 millikelvin, and we use a dilution refrigeration to get the systems that cold.
Tonya Hall: You've had a five-cubic quantum computer available online for two years now, and tens of thousands of people have run millions of experiments on it. So what kinds of questions and problems has it been used for?
Talia Gershon: This is a really great question, and actually when we first put these devices online back in 2016, we had no idea what people were going to do with them. One thing we knew for sure was researchers would use these systems to conduct their own research. We knew that people without a quantum computer that were trying to do quantum computing research would use them for experimentation, for testing theories, for testing, creating new algorithms, for discovering new ways of mitigating errors and things like that, and all that has happened. But what's been really amazing to us is how many just general tech enthusiasts have gotten their hands dirty trying out different things, creating entanglement and real physical systems just for educational purposes and to learn.
Tonya Hall: I know for a fact people have been creating games. Our priorities are in order, Talia... talk about that. I mean what's the point? Why are they building games on quantum?
Talia Gershon: I love that. I love that people are creating games. It's really very cool. There's a few things that I think are really interesting about this. One: Games are a mechanism of learning. When people play games... I remember when I learned to type, I learned to type through games. Games are a great way of learning. We recently put out a game called Hello Quantum, which was originally just in iOS, and either already or very soon we're going to make it available in Android and on the web, where it just allows you to play around with different kinds of gates or operations that change the state of the qubits. So, you can go in and you can play around, and through solving puzzles you can learn about quantum gates and the role they have in quantum computation. It's kind of an easy on ramp into the field. It's kind of lowering the barrier to entry for people. That's one really interesting thing.
Another thing is there's been a lot of ways in which games have gotten software developers into a new field or into a new area. We're hoping that the opportunity to create the first-ever quantum games might attract new kinds of people to the platform who might not have otherwise been that interested in quantum computing.
Tonya Hall: Do you have any data on the types of users that you have? Are they young? Are they old? Are they universities? I mean who's actually accessing the quantum computer?
Talia Gershon: We don't really track our users in that way, but we do know that about 100 or so research papers have been published by people across the academic community who are just using our systems for research purposes, which is really very cool.
We've had lots of people reach out to us. There's over 1,500 universities, different universities, where at least someone from that university has logged on to the system and done experimentation. High school students are across the country and across the world.
So a lot of people are using it to learn, which is really cool. There's been a number of courses that have been taught using the IBM Q experience as part of the educational curricula, and there's been a number of professors who have reached out to us about the ways that they want to use the platform in their classes, and things we could do to enable that. So that's been really exciting.
SEE: IT leader's guide to the future of quantum computing (Tech Pro Research)
Tonya Hall: Tell us about the state of quantum computing today. I mean is it the research experiments that we're talking about, or are we actually solving real-world problems already?
Talia Gershon: It's a really exciting time in the history of quantum computing. I think people don't appreciate how recent it is, that we have the kinds of systems that can be stable online for long periods of time. Two years ago, when we launched the IBM Q experience, that was the first time a real quantum system was made available through the cloud, in any kind of reliable way. And now we have 20 cubits systems accessible by our clients in the IBM Q Network. We have 50 cubits systems under development.
People are starting to begin to translate the kinds of problems that we hope to one day solve on quantum computers onto the actual quantum system. So being able to actually figure out how to represent those problems on a quantum computer, these kinds of problems are being figured out right now. This is all research. So it's a very exciting time.
We're in a period where we like to call getting quantum ready. This means one day we're going to have systems that can outperform the biggest supercomputers in the world. And that's a pretty high barrier to cross, being able to beat supercomputers that people have been developing for decades is going to be a major, a major milestone. But if somebody handed you those systems today, nobody would know how to use them. So there's going to be a period of getting ready, creating the software stacks, figuring out how to mitigate errors, figuring the algorithms out, all these things need to be figured out so that when we have those systems, we can make the best use of them.
SEE: Photos: The world's 25 fastest supercomputers (TechRepublic)
Over the course of the next few years, we think the systems are going to advance, and the algorithms, and all the implementation... the software stacks are going to advance, sort of, in parallel, until we get to the point of reaching a quantum advantage era, where we can actually start using these systems for an advantage, for a computational advantage over the kinds of systems we have now.
Tonya Hall: Explain the relationship then between traditional computers and quantum computers. I mean, will we always still use traditional computers to connect into a quantum computer?
Talia Gershon: Yeah, no, we definitely will. Quantum computers are not going to be replacing regular computers any time soon. You're always going to need a regular computer to program your quantum computer and to control it.
Quantum computers are also not going to be better than classical computers for everything. There are certain classes of problems that quantum computers will be better at solving, but for everything else there's classical computers. We really view the future of quantum computing as a partnership, a close partnership between classical systems and quantum systems.
Tonya Hall: You talked before about qubits. What makes scaling qubits up difficult? What challenges exist when making a 50-qubit machine?
Talia Gershon: Yes, this is a really, really good question because actually patterning the qubits on a chip is not the hard part. We figured out how to pattern trillion and billions of transistors on a chip. It's not the patterning that's the problem. It's keeping the system coherent long enough to do useful computation. It's making it so that controlling one qubit doesn't accidentally control another qubit. It's making it so that after performing so many operations in a row and having the error accumulate, we know how to remove it. It's all of the things associated with actually having the system be robust and reliable and in the presence of a lot of different operations and qubit interactions going on at the same time.
SEE: Meet IBM's bleeding edge of quantum computing (CNET)
Tonya Hall: When will quantum computers become mainstream and for what cases will they be applied?
Talia Gershon: So the beauty of this is that it's research. We're hoping that over the next few years, the first application where quantum computing offers an advantage for computation, those first few use cases will emerge. We think a really high probability area where quantum can have an impact is in chemistry simulations in stimulating nature because nature isn't classical. Nature is quantum mechanical. So using a quantum system to model a quantum system we think is going to be an opportunity for creating an impact with quantum computing.
There's a couple other areas that are emerging from a research point of view, but the beauty is that it's research, and we won't know until we discover it, and that's a really exciting thing.
Tonya Hall: Well, it's very exciting to all of us, and good work. I appreciate you sharing some insight on quantum computing. If somebody wants to connect with you Talia, maybe they want to find out more about the work that you're doing or maybe connect with you personally, how can they do that?
Talia Gershon: So I suggest you reach out on Twitter and you follow me on Twitter, and potentially also follow IBM Research. We got a lot of really interesting things that we're doing in IBM Research overall.
- Quantum computer components: Learn the basics (TechRepublic)
- CES 2019: A peek into IBM's quantum computing plans (TechRepublic)
- What a quantum computer is, and why it needs to be more (ZDNet)
- Quantum computing: Seven truths you need to know (TechRepublic)
- How the industry expects to secure information in a quantum world (ZDNet)
- D-Wave quantum computers: Cheat sheet (TechRepublic)