and as it with the very good classical algorithms can keep up. To what extent the future society? And where are they now? To get the view of the D-Wave in the State of things, we use today with the CEO of the company, Vern Brownell. Mr. Brownell is usually based at the company headquarters in Canada, but it studied the IEEE Spectrum in New York. Vern, podcast welcome. Vern Brownell: Thank you, Rachel. It's nice to be here. Rachel Courtland: from the outset, I wonder whether he could speed up us: D-Wave quantum computation approach is slightly different, which is continued in many research laboratories worldwide. They break was the key for us differences? Vern Brownell: Yes, absolutely. Most of the quantum computation and Shors factorization algorithm talk, it relies on the so-called Puerta Modelo or quantum circuit model. Most of the research, which was — I would say that 90% of research, who has done on the field — was in this calculation model. But there is to build several ways to the quantum computer. One the founder of our original principles of the company is that we decided to use a model called the adiabatic quantum computer, the theory, the most developed with, I think it was in the year 2000, she published an article that Key To The Quantum Industry describes the theory behind it. And in fact, what we have done is to apply this architecture. It is therefore very different from La Modelo. It is not the direct and Shor algorithm uses a different algorithm, some sort of dispute, perhaps to solve, is that this model is smaller than the model of the door or some other models of quantum computation. Rachel Courtland: and you solve these problems. Vern Brownell: So our car is well suited for this, we think, has three main categories of software applications and algorithms today. If we focus some features as standard that is in fact a probability distribution. Recalls especially the Monte Carlo simulation, so most commonly used in finance and modeling risk and stuff. The second category is the optimization. It turns out that the computer architecture natively is an optimization problem. Optimization problems as a traveling salesman problems or logistical, things like: particularly suitable for architecture, and then the third category is in a wide range of machine learning, we believe that this is one of the most exciting things to happen in computing today. Things like depth and technical learning that are ripe for exploration in the area. Rachel Courtland: computers as D-Wave have been for nearly three years on the market. Are purchases of Lockheed-Martin and Google mean for the company? Vern Brownell: Well, for a start, they are great companions. Can not questions for the best customers first by Google and NASA and Lockheed-Martin and so on. So, we are very proud and honored to work with them. Lockheed, as our first customer, their motivations and proposal for the primary use of the machine, is on the software verification and validation, especially in complex systems and flight control systems, in particular. So is the main area Lockheed exploration, even if we have much more in other categories, which we languages. And Google, of course only mentioned machine learning. Their use results in case this is a sample application, a toy, introduced with Google a few years ago, which can detect a player in an image at the bottom. Use learning techniques to do. Rachel Courtland: it seems that there is a great debate about the nature of their quantum computer. One question is how much and whether they soar more the best classical algorithms. We would say that still must be demonstrated by the wave-D speed advantage offer? Vern Brownell: We are in advanced. If you think what we do, we have algorithms that also may run a little more well-known algorithms for the same operation with a classical computer. And when I say classic I find the rest of the computer industry and so on. If you think what does it mean in comparison, there were 60 great years of the development of the material by John von Neumann in 1948 with the ENIAC and all advanced material and all progress say software, all work of the algorithm and everything that's happened, billions of investment, 60 years. They built us this team for 10 years, when we started in 2004 is almost as good, and even a little better maybe sometimes narrow to develop in this ecosystem. And think that great achievements. And our new generation machine, we believe that reference more than one algorithm for some case. We are on the edge of the historical outcome, I think, in computer science, a completely different mechanism and orthogonal layout, design if you, in the computer industry, which in addition to this ecosystem want. Rachel Courtland: another question open, is the question of scale apparently. At the moment sold computer with 512 quantum bits or qubits. What size finally wants to make this team, and what you need to do to get there? Vern Brownell: that is our intention in basically every two years with a new processor, which is actually quadruple the number of qubits. The basic building block for our car, decided, semiconductor processes, to use, there is a wafer, so on an area of fabulous, shopping, said and is integrated into our unit and it's the machine. In 2010, we have published a 128 Qubit. We have the early 2013 a Qubit-512. And we offer more later one Qubit 1000-processor make this year. It's like we get a constant amount of iterative development, both in the learning process and Fab, to develop the processor and interact with our customers. Draft future to integrate feedback and build best processors and better and faster and faster. This is our roadmap, and we see it build no limit to the number of qubits a. There may be physical limitations, to cross us, but simply continue to build our philosophy is, and when we problems on the way to work around those and more. Rachel Courtland: my understanding is that there is a the key ingredients you must have the number of qubits enough capable of a quantum computer, which is all qubits with others tangled cables to extend the same quantum state in parts. I know there was a discussion on entangled qubits as D-wave. Where such research is now, and it responds to this criticism? Vern Brownell: is a good question. First of all, I'm no physicist, Rachel, as you know. I am engineer in the background, but if I by osmosis to understand and work with many physicists in the years, I think it's safe to say that it is not clear is what role of involvement in these calculations of what we do. It is certainly a useful indicator if, at the lowest level, it a quantum computer or not. Is an element of the list of things that seem to be scientists, and that's why we have developed a series of previous experiences in the past year, the 8-Qubit entanglement showed in the height. We have, as I said, is to develop the participation of science and progress, but also economically meaningful processors. It is more and more difficult, such as rock climbing, because performance, a processor models, which are necessary for the performance or the expected to charge to more complex and difficult. Sometimes it is a quantum computer model the skills with classic how to do. But we continue this investment to make, and it is quite clear that entanglement on level 8 Qubit and beyond. We will continue with the community to work and try, continued the series at larger scales. Rachel Courtland: is, where it is intended that the d-wave will be in the next few years? Vern Brownell: Well, as I said, the Roadmap basically should obviously these more powerful processors. It's interesting: If you are running this criteria and benchmarks to do still careful and difficult, when comparing two conventional systems, but it is very difficult to compare a conventional system a quantum system. But I think that in the coming years we will be a very clear discrimination against all classical system. So, our processors will become quality. We have other software tools. We offer this service in the cloud, that people can buy without a large D-wave machine and install it in your datacenter or in our data center hosted. I think one of the things that I try to on the D-Wave and D-Wave has in its DNA, is the goal of the advancement of science, but science and technology are better, it is possible to translate. It is very hard to do, but I think it is very valuable, and we will continue on this way. Rachel Courtland: Vern, please visit today. Vern Brownell: Thank you, Rachel. Rachel Courtland: we language with Vern Brownell, the CEO of D-Wave, work from your company in quantum computing. IEEE Spectrum Techwise are Rachel Courtland talks. This interview was recorded Thursday, March 20, 2014. Sound engineer: Francesco Ferorelli photo: D-Wave Systems Learn more. Rachel Courtland: Hi, I'm Rachel Courtland IEEE Spectrum Techwise talks. twenty years of the mathematician Peter Shor found what the killer app is called by quantum computation: the ability to factor very large numbers much, much faster than a conventional computer. Since the research in the construction of quantum computers really took off. It is very slow going. A big problem is when creating a new quantum system seemingly fairly simple operations, such as the number 15 3 and 5 numbers factoring. D - Wave, a company headquartered in Burnaby, British Colombia, has tried to speed things up. But it was not completely smooth sailing. In the year 2010 in an edition of special profiles, win and lose the technological projects last year. . At the moment told external experts ourselves, that it was not clear, exactly, as it was D-Wave's quantum systems, and how big it could become. Since then, D-Wave has reached a remarkable series of victories. In 2011, the company. ,,.