What is a quantum computer? Difficult, but feasible question. And how does it work and what is it for? Here it is more difficult. Our Microsoft Research team within the Quantum Project (QuArC) is developing a quantum future. And they do it. In this article you will find an interview with a member of the development team - Krista Svor. She will tell about qubits and -273 ° C. More under the cut!
When it comes to quantum computing, and you nodded thoughtfully in response, although in fact there’s no doubt, many people are also perplexed. Some of the smartest people on the planet admit that they, too, do not understand what it is. Fortunately, other well-known figures, such as Dr. Krista Svor, the chief research officer for Microsoft Quantum or QuArC at Microsoft Research in Redmond, really understand the essence of quantum computing and are working hard to make this technology a reality.
')
Today, Dr. Svor will tell us about his passion for quantum algorithms and their potential advantages in terms of solving some of the most complex world problems. She will explain why Microsoft’s topological quantum bit or “qubit” was a revolutionary discovery in the field of quantum computing, and also tells that despite the fact that the quits themselves “live” in cryogenic-vacuum installations at a temperature close to absolute zero, The researchers of the quantum world continue to work in their cozy offices, using a computer program created on the basis of Schrodinger's cat.
So, Krista. Your research focuses on studying quantum algorithms. Tell us what the peculiarity of quantum algorithms is and why you are attracted to this particular area of ​​research.
In my opinion, quantum algorithms can really help us solve the problems that digital computers simply cannot cope with. Quanta, or quantum computing in general, offer us tremendous opportunities. So for me, quantum algorithms are a way to solve various problems using special computers.
Algorithms are scripts that are executed by a computer. And so the development of quantum algorithms that can solve real-world problems and have a real significant impact is a big deal, and this is my big passion. I want to see how we can change our world, change our ideas about clean energy or the production of organic food. I mean tasks that affect all of humanity. We can solve some of them, or at least do our best to solve them, with the help of quantum algorithms and quantum computers.
At what point did you realize that quantum algorithms are exactly what you really want to do?
Yes, good question! In college, I studied math. In our first year at university, we had a seminar led by Andrew Wiles, a famous mathematician. He proved the great theorem of Fermat. This seminar was devoted to cryptography, and so at this seminar he shared with us the idea that there must be some kind of computing model capable of hacking the RSA cryptosystem, which we use today everywhere, with the help of factoring. He told us that this model of computing, namely the model of quantum computing, can actually crack this cryptosystem. It was then that I was inspired by the idea that there is another model of computation.
I was also very interested in computer science, so I wanted to understand how this model is used in practice. That is, how to write such an algorithm for computing? How to create a programming language and software stack, then to implement this algorithm to work on real equipment? These questions marked the beginning of many years of research and my desire to engage in quantum computing.
So, we are counting on a rather high level of technical training of our students. Perhaps we can not call this podcast "Quanta for Dummies."
Yes.
But I think that a small excursion will not be superfluous. So, what are quantum computations and how do they differ from classical computations? We have two minutes to do this. Tell that.
So, quantum computing is based on the principles of quantum mechanics. And these principles are fundamentally different from the classical ones. Superposition, quantum entanglement, interference, you met these terms if you listened to Feynman’s lectures, Richard Feynman’s lectures, or studied Einstein’s notes and articles.
For example, quantum entanglement is a very interesting phenomenon, in which the two systems are interdependent, and they retain their interdependence even if they are in different universes. So when any changes occur in one system, they instantly manifest in another. This is just incredible. Or here is the phenomenon of superposition. If I have a quantum computer, I use superposition to store information on it. That is, bits are not used, as in a classic computer, a classic computer is, of course, just binary. It turns on and off. It works like a switch.
As for the quantum computer, I would compare it with the operation of the smooth controller. Your information is stored in a quantum state. And in this quantum state, 0 and 1 can coexist simultaneously. This is a linear combination of zero and one. And this makes scaling possible and allows you to achieve a truly amazing exponential increase in performance using quantum computing.
You lead the Microsoft QuArC team at Quantum Architecture and Computation (Quantum Architecture and Computing) at Microsoft Research in Redmond. Tell us about your team, partnerships and your vision for quantum computing.
Yes, we created QuArC, I think, seven years ago. We began to think about what we actually want from the quantum computer. He has great prospects, great potential. But how to put all this into practice? How will we program? How to implement the algorithm at the hardware level? You remember that the principles here are completely different? Systems are fundamentally different.
For this we need classic computers, different from those that were created earlier, besides, actually, a quantum computer and a quantum system, which also have never been created before. And we focused on not only finding answers to these questions, but also making these technologies more accessible.
So, you just managed it. Tell us about the development package that you recently released.
Yes. We are so excited. We have released the Quantum Development Kit. This kit, the Quantum Development Kit, really helps make quantum computing more accessible to developers, quantum technology enthusiasts, students, scientists, researchers, and all those interested in learning how to use this unique resource.
Of course, the kit includes a new specialized programming language for quantum computing. We called this language Q #. And embedded Q # in Visual Studio. Thus, you not only get this new programming language, but you can use with it all the additional benefits of Visual Studio - all these great tools, such as debugging, code highlighting, syntax highlighting and hints, that pop up when you hover over functions and operations. This approach makes these technologies much more accessible and, I hope, reduces the “entrance threshold” to the world of quantum computing.
Do I need to be an expert in quantum mechanics or quantum physics to understand all this and start programming for such systems?
You know, we very much hope that this is not necessary. If we focused only on developers who understand quantum mechanics, then this would be a very small group of specialists. And we really want to present quantum computing for a wide audience, to make them accessible to all. So the idea is that you don't need to know quantum mechanics.
But I myself am from this field, I am an expert in computer technology. I am not a physicist. I did not study quantum mechanics in college. I come to this from the point of view of an expert in the field of computer technology, who wants to use it to change the approach to the organization of calculations. We hope that more and more people will begin to discover this new area for themselves and begin to develop algorithms and applications in this environment. But quantum mechanics is not necessary to know.
And this is great because many people can immediately abandon the idea of ​​entering the world of quantum computing, if the input threshold is too high for them, so ... How much Q #, this quantum programming language, is different from other programming languages, and does Is it currently on real quantum computers or only on emulators?
Q # is really designed for quantum computers and the computing model used on a quantum computer. But he inherited a lot from classical programming languages. It is important to remember that a quantum computer is a hybrid device. It looks like a coprocessor or accelerator. That is, you write most of your application and the wrapper in a standard language, and then call a program developed in the Q # language, which, in turn, runs on a quantum accelerator.
We developed the Q # language, relying on such a concept. The basic idea is to use Q # to simplify things such as superposition, entanglement and interference, in order to develop your quantum algorithm and ensure its development. The functionality involved in quantum algorithms is fundamentally different from those we use in classical algorithms. And with the help of Q #, we bring this all to the fore, simplifying the task for the developer.
One of the main problems in quantum computing is the instability of qubits, and this problem, in your opinion, should be solved by the concept of topological qubit. Or already decides ...
Yes.
So, what is a topological qubit and how can it help in quantum computing?
True, in quantum computing we use qubits. Qbits are a way to store information in a quantum computer. These are "quantum bits". Shortly, “qubits”. So, it is important to understand that these qubits constantly interact with their environment. This is not a good property for quantum computing. We need them to remain in a vacuum, so that they are isolated from thermal noise and other noise. And when they start to get confused with their surroundings, it means that they become "noisy." This entanglement with the environment leads to the loss of information in the qubit. This should not be allowed. The system needs to be isolated.
There are various ways to isolate the system. But Microsoft’s approach is truly unique here, and it provides maximum scalability. We must develop a method for storing and processing information on a quantum computer. And while ensuring higher reliability. I mean, right now, at the concept development stage. Imagine that you want to build a skyscraper, and I have a brick. And we all believe that the bricks are pretty durable. But you will not try to build a skyscraper of bricks?
Right.
It is unrealistically difficult. That is, you are likely to abandon this venture. For this it is better to use steel. So, steel is better suited for the construction of a skyscraper, and this material will allow you to quickly scale the project. This design will be much more resistant to changes, natural disasters and other external influences.
So, imagine that our qubit is a steel qubit. And our system based on steel qubits, topological qubits, as we call them, will scale much easier, and this will require less effort, less additional resources, less additional costs, less extra time, and as a result you will get more computing capabilities. more qubits. That is, you can do more with less effort. It will be possible to perform more calculations using fewer qubits, since they are like steel. They are very, very strong.
So you want to say that this is a different type of qubit?
Right.
And this is Microsoft Research's know-how?
Yes. Research by Mikhail Friedman, who joined MSR in the late 90s, and his ideas formed the basis of this qubit, he worked on them together with Alexey Kitaev, who at that time worked on the Microsoft team, as well as with other colleagues and partners in several universities. The ideas that gave rise to the development of this qubit were actually born at Microsoft, in particular at Microsoft Research.
That is, here they were transferred to the conceptual plane.
Yes exactly.
Let's talk about the quantum computing system. A quantum computer does not work by itself, it needs a certain ecosystem. Tell the listeners a little about what should happen before the quantum computer can start working in our limited world.
This is about the quantum ecosystem, and Microsoft has its own unique approach, our approach is to build a comprehensive scalable quantum system. It is not enough just to build a quantum computer that will give us 50 or 100 qubits. We want a quantum computer to scale thousands of qubits, tens of thousands, hundreds of thousands of qubits and more. This will give us truly unique opportunities and help solve the most complex world problems.
So, if we are talking about an ecosystem, what does it consist of? First, we need a qubit for storing and processing information using the principles of quantum mechanics. This qubit is stored in our system at a temperature measured in millikelvins. To make it clearer, this is very close to absolute zero. That is, first of all, you must place the entire system in a cryogenic plant to keep qubits cold, protecting qubits from noise, from interaction and entanglement with the medium. In addition, we need a system or stack if we want our quantum computer to work, right? That is, we need programming tools.
So, we are developing quantum algorithms, and I do not know how the rest of the listeners, but I prefer to create quantum algorithms at room temperature. And I will need to interact with this system, which functions at absolute zero, which is about minus 459 degrees Fahrenheit. Somehow I need to transfer this program, this controlled information, to a quantum chip. This means that the system must have a management area. And we will have to run quantum computations, subroutines, if you will, from under the classical program. That is, without a classic computer still can not do. And this computer should work, say, at minus 270 degrees Celsius.
Thus, we must design and create not only a quantum chip, but this classic computer, which is not at all like any other classical computer that we have ever developed. It should work at temperatures at which we never even tried to run classic computers. So we design it too. We develop the quantum chip itself and all the software for bundling all the components. Then you will need to compile the program for this system. Therefore, we are developing not only an integrated hardware stack, but also an integrated software stack.
That is, you need to solve an interdisciplinary task.
Right. Of course, another key and unique feature of our quantum program, Microsoft and Microsoft Research, is our team. We have brought together experts in various fields in a wonderful international team. These are not only physics scientists. Experts in computer technology are also needed. Looking for engineers specializing in electrical engineering. Without specialists in cryogenic technology, too, can not do. And developers, software engineers ... You can continue for a long time.
And we managed to assemble an amazing “dream team”, including specialists from more than 10 countries of the world. Together we are working on a complex quantum system, taking into account absolutely all of its aspects. The team included real professionals in their field. Do you know, for example, that Michael Friedman, who helped launch the quantum program itself at Microsoft, is the Fields Prize in Mathematics? Our team works founder of the company Cray Computing. We have a leading architect from Intel. We work with specialists in the field of experimental physics from the professorial staff of Harvard and Delft, as well as scientists who have received awards for various achievements in computational physics. All these wonderful people work together on this task, they burn themselves and inspire others.
Including you.
Oh yeah, thanks.
Let's go to the cream on the cake. , . - , . ? ?
, , . , , . , , . , , 1000 10 000 . « ».
, , Microsoft, . . , . , , 1 000 10 000 .
.
Yes it is.
, ?
, , , .
.
, .
— , ?
, . .
, .
Of course. , Microsoft Research 11 , , . Bing -, . , . , , , .
6 , , . , . , . , , , , , .
, , , . . . , , , , , . , , , . .
. , , , , , . , , , .
Right. , 1994 . , . ? , , . : «, !», ? . .
, , . Microsoft Research . , , , , . . , , RSA. , , Microsoft Research , , , , . , , , , , RSA .
.
, . , , RSA, , , , . , . .
…
Right.
?
Right. , , , , - . , . , . , , , .
.
.
. , , , , . ?
, , , - , , - . , . . , , , , . . , , . . « » .
, . , . , , , , .
Microsoft Research? ?
, , , Microsoft Research, , Microsoft Research . . . . Microsoft.
, , , . ; , , , , , , .
, , , , . , . , .
, … , . , , Microsoft — - , .
, ! , . , - , . . . , . . , , .
, - .
, .
, ?
, . - .
, .
Right.
. , , , , - , , . , , , . ? . , , - ? - , .
, ? - , .
. . .
Yes.
, . , , , - , , - . , . ? , , NP-hard, , .
, , , , , , . . . , ?
?
, . , . , , . , , ? , , , , .
, , , . , , , , . . , . . , , , .
…
, . , , - . . But…
.
, , , . , , , . , . , , , - , .
, , , , , , , . . . , , , ?
.
, .
, , -, - ?
, — . , ? , , , , , .
, . , , . … , , … , , … , , .
. , « ». , , .
!
, , .
, , . .
.
Yes. For sure.
Microsoft , microsoft.com/research .
Ps , Qubit (Quantum Daily) Telegram.