Quantum computers and prospects for their development in Russia

46
Quantum computers and prospects for their development in Russia

Nowadays, technologies are developing by leaps and bounds. Progress has accelerated so much that sometimes people simply cannot keep up with the rapid changes.

At the same time, many experts predict a real breakthrough in science and technology when quantum computers, a device that uses the principles of quantum mechanics to process information, begin to be used en masse to solve applied and other problems.



Unlike classical computers, where information is encoded in bits (either "0" or "1"), quantum computers use qubits, which can be in a superposition of "0" and "1" states simultaneously. This allows calculations to be performed in parallel, significantly increasing the speed of solving certain problems.

The basis of the operation of such machines are such phenomena as superposition, entanglement and quantum tunneling. Superposition, as mentioned above, allows a qubit to be in several states at the same time, entanglement provides instant communication between qubits, and quantum tunneling helps to overcome energy barriers when performing calculations.

Today, leading companies in the world, such as IBM, Google, D-Wave, and others, are actively developing quantum computers.
At the same time, today these devices can be conditionally divided into two categories: universal quantum computers and specialized quantum devices.

General-purpose quantum computers, such as IBM Quantum System One, are designed to perform a wide range of tasks, including optimization, modeling complex chemical processes, and encryption. In turn, specialized quantum devices, such as systems from D-Wave, are aimed at solving narrow problems, such as logistics optimization or big data analysis.

Quantum computers have already reached quantum supremacy, the point at which they begin to outperform classical supercomputers at performing specific tasks. An example of this was Google's achievement with its Sycamore quantum processor in 2019.

Quantum computers have a number of advantages that make them promising for solving problems that classical computers cannot. In particular, thanks to the superposition and entanglement of qubits, these devices can simultaneously process huge amounts of data. They are able to find optimal solutions to problems with many variables, such as logistics, financial modeling, and drug development.

In addition, quantum computers are ideal for studying chemical and physical processes at the atomic level, and quantum algorithms such as Shor's algorithm are capable of breaking any existing encryption system, but also help develop new ones that are resistant to attacks.

It is worth noting that Russia also has its own quantum computer. The 50-qubit device created by the Joint Laboratory of the Lebedev Physical Institute of the Russian Academy of Sciences and the Russian Quantum Center (RQC) has become an important step in the development of relevant technologies in our country.

The device is built on the basis of qubits implemented in superconducting circuits, which have high stability and minimal error levels. This allows quantum calculations to be carried out with greater accuracy.

The device is equipped with a sophisticated control and cooling system that maintains superconductivity at temperatures close to absolute zero (-273 °C).

50 qubits is a significant achievement, as such devices are approaching the threshold of “quantum supremacy.” At this point, such a computer can perform certain tasks that are practically inaccessible to classical supercomputers.

It is worth admitting that we are still far behind our competitors. For example, IBM presented its Quantum Condor with 2023 qubits in 433. Meanwhile, the development of a 50-qubit quantum computer is only the first step.

In the coming years, Russian engineers plan to increase the number of qubits, improve the stability of quantum states and reduce the level of errors in calculations. In addition, work is underway to create application software that will allow integrating quantum technologies into existing information systems.

Thus, the developers set themselves the goal of adapting the computer to solve applied problems in various fields, including: cryptography, materials science, logistics and finance.

Overall, the future of quantum computing in Russia looks promising, despite existing challenges such as Western sanctions. Government support, joint efforts of scientists and businesses, and a high level of training of specialists create the basis for our country to become one of the leaders in the field of these technologies.



46 comments
Information
Dear reader, to leave comments on the publication, you must sign in.
  1. +14
    17 January 2025 12: 23
    Well, everything is stated briefly, clearly and understandably. Only a completely stupid person will not understand qubits, superpositions and quantum tunneling. And the fact that "qubits can be in several states at the same time" is clear to any layman. But I understood one thing well - we are again in the role of laggards.
    1. -2
      17 January 2025 12: 29
      Only panic-mongers and potential traitors think like that. Secret quantum computers based on new physical principles and having no analogues in the world have been developed in the depths of secret underground laboratories for a long time. Calculations for secret projects are in full swing, you will see everything for yourself soon - there is not much time left to wait.
    2. +3
      17 January 2025 12: 34
      The difference between 433 and 50-qubitamm is solely in the size and cost of the installation. The latter option, in my opinion, is more profitable from the point of view of debugging technologies in the absence of applications for solving specific problems.
    3. +3
      17 January 2025 12: 52
      "But I understood one thing well - we are again in the role of laggards."
      The one who is lagging behind can catch up, but the one who is late cannot.
    4. AB
      +4
      17 January 2025 13: 06
      I personally thought that we only had Soviet developments in this area, but they were apparently being conducted. Therefore, for me it is a very pleasant surprise that we actually have quantum computers! Well done!
    5. +2
      17 January 2025 13: 06
      And yet it is curious how it works. In several states - it is as if. Or rather, they do not have a certain one. It is unclear how entanglement is used - after all, there is also a probability calculated by formulas, and not a 100% result. Probably, at one time, it was necessary to study physics
      1. +1
        17 January 2025 14: 34
        Still, it's curious how it works.

        You won't understand. And physics won't help. It's not even about the computer itself, it's about Mathematics.
        But the algorithms for quantum computers use a different ARITHMETIC. There are other basic operations, not + and -.
        To put it simply, a quantum computer is an analog computer.
        Like a logarithmic ruler or bradis tables. There was also an advantage there (according to tables, trigonometric functions are calculated instantly laughing )
        1. +1
          17 January 2025 14: 50
          When you press a key, you don't really think about how it happens (I'm talking about the result). What sine and what cosine does it relate to the theory of relativity.
  2. -1
    17 January 2025 12: 42
    Quantum computers resemble neural networks. They have been made for a long time. And their practical benefits are about the same, and they are mostly for the well-being of developers. What have neural networks succeeded in? Inventing new drugs for cancer or covid? Maybe they build cunning financial strategies? Will they lead us out of the crisis? Will they win the war? Yes, they do more covers of songs about sewage workers and stock pictures and enrich Huang request
    1. +2
      17 January 2025 13: 21
      If you personally haven't heard or haven't made a summary of where and how neural network technologies are used now, then don't immediately pretend that they haven't helped anywhere. A lot has started up now and is based on neural networks' developments. I'm not even saying that the neural network sphere itself has been used for a long time. It's still well before the peak of popularity of drawing programs.
      The same goes for quantum computing. There are already computers using some quantum computing technologies. Yes, there is no full-fledged purely quantum circuit yet. But it is not needed yet. They will be needed for very specific calculations that need to be greatly accelerated. This is just around the corner. I am not even talking about the fact that around the main direction of development there are a lot of related technologies and solutions that are already being used in various areas, for example in encryption.
      You probably want something like a switch, so that tomorrow on TV the whole world would say, "eureka" and something like that would appear and immediately here is a new era. Well, it won't be like that. But like you see it now. You won't notice how it has already happened, and part of your life depends on systems based on the new.
      1. +1
        17 January 2025 14: 18
        If you personally haven’t heard or haven’t made a summary of where and how neural network technologies are currently used, then you shouldn’t immediately pretend that they haven’t helped anywhere.
        I thought the examples I missed would appear now, but no
        You probably want something like a switch, so that tomorrow on TV the whole world will hear “eureka” and something like that will appear and immediately there will be a new era.

        Because not everything at once. You need to buy shares of whom you need to buy right away, and the results will come, but later.
        I remember 20 years ago they predicted that the atomic force microscope would create nanomachines tomorrow that would repair human cells, and in 2004 they wrote that the most promising professions would be related to nanotechnology. And somehow they gradually moved away from the topic, Chubais probably broke world science. Now instead of nanomechanisms there are neural networks that are about to revolutionize the drawing of frames and change life for the better
        1. 0
          17 January 2025 14: 48
          I thought the examples I missed would appear now, but no

          Catch
        2. +1
          17 January 2025 14: 57
          It’s just that you, like the media, like a huge crowd of people, don’t understand two aspects.
          As for neurons, this is not a miraculous discovery like the periodic table or an invention like the internal combustion engine.
          This is a clear, fairly simple and long-known tool in IT.
          It is very useful when you have huge amounts of information.
          If these volumes are not there, then there are no neurons, and there are not even tasks for them.
          Please note that you currently have more information on your phone than the USSR Academy of Sciences Computing Center of 89. Yes, I worked there too. laughing
          About quantum computers, quantum computers in the foreseeable and even unforeseeable future will not replace conventional computers. Because they are applicable only for a very narrow range of tasks. You can consider that they are analog computers. That is, outside this range of tasks, they are not the future but the past.
          1. +1
            17 January 2025 15: 33
            This is a clear, fairly simple and long-known tool in IT.
            I know that the data processing method is just hyped up. Handwriting or face recognition - this has been known for a long time
            It is very useful when you have huge amounts of information.
            the previous mega-super-advanced IT was big data. Now companies have a lot of data, developers have a lot of money, that's all
            Because they are only applicable to a very narrow range of tasks.
            probably will greatly advance string calculations. Maybe after 30 years they will find a method to connect strings with reality, but for now they have been calculating spherical twelve-dimensional macaroni in a vacuum for 40 years
          2. 0
            2 February 2025 00: 44
            About quantum computers, quantum computers in the foreseeable and even unforeseeable future will not replace conventional computers
            It is clear that unlike classical computers, quantum systems do not have a cloning operation (the fundamental prohibition on cloning a quantum state), only a negation operation remains. But that is not the main thing. Will it be possible to play on them? Will Atomic Heart run at maximum settings?
        3. 0
          18 January 2025 14: 19
          I thought the examples I missed would appear now, but no

          I am always surprised by where such a thirst for everything to be put on a plate and described to him comes from. Look for yourself. I have nothing better to do than waste my time on unnecessary movements to prove something to a no-name on the Internet. I will say, I will say, so that it does not seem as if your opinion is the only one, but I will try for your sake... if you please. If you are too lazy to dive in yourself, then even my actions will not be of any use. Tomorrow you will again speak the same old words.
          1. -3
            18 January 2025 17: 56
            I chased you for three days to tell you how indifferent I am to you. laughing
            1. 0
              20 January 2025 10: 42
              Of course not, simple communication is not difficult for me, but conducting searches, if you please. I don't want to. Without me. Yourself, with your hands, with the help of a keyboard. You can write here, so it won't be more difficult in Google.
              1. -2
                20 January 2025 12: 45
                And you know, I didn't just stupidly search the Internet, I also tried to use it to my advantage. So, fairy tales of professional storytellers don't work for me. All this should be divided by at least ten, and more likely by a hundred.
      2. 0
        17 January 2025 14: 44
        They will be needed for very specific calculations that need to be greatly accelerated. This is not far off.

        Of course. But it seems that no one, including the author of the article, understands this.
        While it is true that the calculations in which they are used are EXTREMELY specific.
    2. +1
      17 January 2025 14: 42
      What have neural networks succeeded in?

      For example, in translation from one language to another.
      Or in search of tumors on MRI.
      And yes, neural networks search for targets on satellite images.
      Neurons are searching for criminals through cameras.
      Yes, we can go on for a long time.
      1. -1
        17 January 2025 15: 34
        For example, in translation from one language to another.
        Or in search of tumors on MRI.
        doesn't look like a revolution
        And yes, neural networks search for targets on satellite images.
        at least twenty years ago, I was told this in a lecture when Rusnano didn't exist yet
        1. +2
          17 January 2025 16: 03
          In theory, a quantum computer created for a certain computational model of a neural network (or vice versa) could be downright insanely fast (like any answer in a split second).
          In practice, a hundred mega-video cards costing tens of millions of rubles each will do about the same thing, but will be more compact and IT ALREADY EXISTS AND WORKS.

          Again, even one of those video cards with a memory capacity of ~50 GB can already do a lot.
          Without requiring nitrogen, special conditions (maximum a server rack with air conditioning and even then not always, more often it is DESIRABLE but not necessary).
          1. 0
            17 January 2025 16: 49
            It remains to be seen - what kind of tasks are these? I heard that supercomputers do everything, but somehow they didn't manage to come up with something large-scale for them. That's why they are distributed for sharing all sorts of small tasks like calculating docking or something else that can be done on your own computer.
            And neurons are 99% pictures and music
            1. +1
              17 January 2025 17: 58
              Quote from alexoff
              And neurons are 99% pictures and music

              That's not it.
              This (under the hood) is 100% vectors\matrices. laughing
              And the text there, or something else, is not the point.
              1. +1
                17 January 2025 19: 13
                And all computers are zeros and ones under the hood request
                1. 0
                  17 January 2025 20: 31
                  Just why this "pictures and music" thing?
                  (although yes, it ends up being used to generate porn and songs about shit trucks wassat )
                  1. +1
                    17 January 2025 20: 43
                    This is what all this is for. Zeros, ones and vectors are of no use to anyone, and the point of neural networks... Well, like the Internet as a whole in its most degrading manifestation, and expectations are generated by journalists and believers in science - there is nowhere to go. I am a chemist, I know many calculators, over 20 years the productivity of computers has probably increased a thousand times, and the point of all these calculations is 1000x0=0 negative
                    1. +2
                      17 January 2025 21: 16
                      Quote from alexoff
                      what's the point of neural networks...

                      How to say.
                      They perform translation, text recognition and the like quite well.
                      Translation of set X into set Y, roughly speaking).
                      But building a "mega-intelligence" from them, as in science fiction, is stupidly expensive, well, it's not like "networks are getting smarter", it's "they are getting bigger", as a result, more machine resources are required.

                      and expectations are generated by journalists and believers in science - there is nowhere to go.

                      HA! Exactly.
                      And this suggests that history is moving in a spiral.
                      There was already "Winter AI".
                      Only then it wasn’t on neural networks, but that doesn’t change the essence.
                      A great start, everyone is delighted and even squealing, money is flowing.
                      And then they couldn't meet expectations.
                      And everything fizzled out, the interest disappeared. The money too.
                      1. +1
                        17 January 2025 22: 29
                        They perform translation, text recognition and the like quite well.
                        Well, it was there before, when we didn't even know about neural networks. Google Translate is a hundred years old, I was able to digitize a Japanese paper patent for translation 10 years ago. Image recognition? Like a student showed on her mobile phone 7 years ago how it recognizes a person and puts an octopus on her shoulder, which hugs and kisses her on the cheek, just like an online animation! The Japanese taught the computer to sing songs on Vocaloid about 20 years ago, and now anyone can do it following the instructions.
                        But when it comes down to it - there's nothing. Actually, the Internet, when it comes down to it, is also nothing - here for money, here there's nothing, and this site was closed and there's not even a trace, the content files were not saved in the web archive.

                        And the funniest thing with neural networks is when you start installing them. It feels like all sorts of ComfyUI were made under something, no instructions or readme, no ready-made builds, only the same users write where to throw what, and in different ways.
                      2. +1
                        17 January 2025 23: 46
                        Quote from alexoff
                        it was there before, when we didn't even know about neural networks

                        a) there were problems with semantics and it was expensive (hordes of linguists and the like).
                        And then a pack of video cards, a bunch of data and "magic".
                        b)
                        Google Translate is a hundred years old, I was able to digitize a Japanese paper patent for translation about 10 years ago.

                        Well, neural networks as such appeared back in the 60s (if not 50s).
                        The first "neural network larva" was crap for recognizing postal codes. Yes
                        Well, in general they were "somewhere there" for a long time, but the recent hype came so suddenly that it seems like they came out of nowhere.

                        Like a student showed it on her mobile phone about 7 years ago

                        Well, it depends on how they did it. It could have been done earlier with OpenCV, but it's not a fact that a phone would have been able to do it (but a laptop in 2006 could have put glasses on its face, for example).
                        The major "milestones" from relatively recent times are acceptable text models, and generation of pornographic (crossed out) images. laughing

                        And the funniest thing with neural networks is when you start installing them. It feels like all sorts of ComfyUI were made under something, no instructions or readme, no ready-made builds, only the same users write where to throw what, and in different ways.


                        Well, if there is no readme, then it is not necessary. Let's skip it. wink
                        A project without a readme with installation instructions is a bad sign.
                      3. +2
                        18 January 2025 00: 04
                        there were problems with semantics and it was expensive (hordes of linguists and the like).
                        And then a pack of video cards, a bunch of data and "magic".

                        Yes, already in 2014, without any pack of video cards, you point your mobile at the inscription and it translates for you. There were probably problems in the nineties, in the mid-noughties, online translators translated English articles more or less.
                        By the way, neural networks have not yet learned to write from Russian to English. You read and understand - well, they don't write like that in English-speaking countries, a three-line sentence with two adverbial participial phrases lol
                        but the recent hype came so suddenly that it seems like they came out of nowhere.
                        Yes, it happens all the time, it's just not as loud. The most striking example is bacteriophages instead of antibiotics, revolution! Then you open nature for 2003 - the same article, they promised a revolution back then. And what's the result? And in the end, a Soviet drug for dysentery has been on sale since the time of Comrade Stalin, someone apparently knocked out money for financing, attracted journalists.
                        Well, a lot of things remained PR. Graphene electronics, a space elevator made of nanotubes, high-temperature superconductors. One step into orbit! But when it came down to it, they simply finished off Musk's quite ordinary rockets, just go ahead and do it, no miracles.
                        Well, if there is no readme, then it is not necessary. Let's skip it. wink
                        A project without a readme with installation instructions is a bad sign.
                        all the image generators are like from Erie floppy disks and punch cards. I didn't think that such crap could exist these days. And who wants user-friendly - paid sites have done everything. One comrade told me that he really struggled for one of these, everything was done in one place fool
            2. 0
              18 January 2025 01: 40
              Well, I took part of the description of the models presented there from huggingface and translated "Audio-Text-to-Text, Image-Text-to-Text, Visual Q&A, Document Q&A, Video-Text-to-Text, Any-to-Any, Computer Vision, Depth Estimation, Image Classification, Object Detection, Image Segmentation, Text-to-Image, Image-to-Text, Image-to-Image, Image-to-Video, Unconditional Image Generation, Video Classification, Text-to-Video, Untrained Image Classification, Mask Generation, Untrained Object Detection, Text-to-3D, Image-to-3D, Image Feature Extraction, Keypoint Detection, Natural Language Processing, Text Classification, Token Classification, Table Q&A, Q&A, Untrained Classification, Translation, Summarizing, Feature Extraction, Text Generation, Text Generation from Text, Mask Filling, Sentence Similarity, Audio, Text-to-Speech, Text-to-Audio, Automatic Speech Recognition, Audio-to-Audio, Audio Classification, Voice Activity Detection, Tabular Data, Tabular Data Classification, Regression on Tabular Data, Time Series Forecasting, Reinforcement Learning, Robotics, Other, Graph Machine Learning."
              1. 0
                18 January 2025 02: 01
                Cool, what's new and revolutionary here? Text and speech recognition is generally the last century. There is creeping progress and that's it, and for the most part it doesn't go beyond the screen.
                All these neural networks end pretty quickly when you realize that it is impossible to depict a handshake, the interlocutor from the neural network is approximately like from Google, and the data processing with prediction has the accuracy of a coin toss. Perhaps the same type of introduction to a scientific article can be rewritten in other words so that the anti-plagiarism does not decide that we are stealing from ourselves.
            3. 0
              10 February 2025 01: 00
              It remains to be seen what these tasks are?
              Problems from number theory and, as a consequence, entailing cryptography. The use of the capabilities of the CC will allow us to reduce asymmetric cryptography to nothing, and then we can work on symmetric cryptography. And this is a very tasty morsel for many special services and government agencies.
        2. 0
          20 January 2025 13: 57
          doesn't look like a revolution

          at least twenty years ago, I was told this in a lecture


          Well, that's what I wrote.
          This is a clear, fairly simple and long-known tool in IT.


          It's true that they most likely told you something different during lectures.
          1. 0
            20 January 2025 15: 54
            The lectures told how a neural network was adapted to search for tanks in the forest for scientific purposes - constructing statistical routes of mice in a maze, to study a cure for dementia.
      2. 0
        2 February 2025 00: 45
        Also in students' essays and for interviews
  3. +1
    17 January 2025 13: 02
    I remember about 5 years ago they were promoting our QUANTUM connection. That it was huge and short-range, but impossible to hack. And then silence.
    Now Quantum Computers. Also wow. What they do, though, is unclear.

    but I read in the news that from January 1st they will charge our scientists money for using a supercomputer (or ... computers).
    Maybe it's somehow related?
    1. +1
      17 January 2025 14: 36
      Now Quantum Computers. Also wow. What they do, though, is unclear.

      A non-specialist cannot understand.
      Of course, I can write an article, but I don’t want to write about something that I don’t fully understand myself.
      But no one fully understands it. laughing
      So for information between quantum communication and quantum computer the only thing in common is the word quantum.
      1. +1
        17 January 2025 16: 10
        Quote: bk316
        So for information between quantum communication and quantum computer the only thing in common is the word quantum.

        Not at all.
        The same as with theoretical physics (strings and all that with relativity).
        Something (abstract laboratory models) even works, sometimes, and there are even rumors that SOMEONE UNDERSTANDS THIS CRAP (but official science stubbornly denies it laughing ).
        But no more. And so (let's say with physics) for more than half a century.
        You can tinker around with something there and retire. Honorably.
        What they all have in common is that if it’s a scam, you won’t be caught red-handed. wassat
      2. 0
        2 February 2025 00: 51
        So for information between quantum communication and quantum computer the only thing in common is the word quantum.
        In both cases, the phenomenon of quantum entanglement (a clumsy translation into Russian, more correctly, connectivity or confusion) between quantum particles is used for work.
    2. 0
      2 February 2025 00: 48
      Meanwhile, Toshiba quietly sells quantum communication systems. Searched for in the search freely.
  4. +1
    17 January 2025 15: 06
    "Nowadays, technology is developing by leaps and bounds. Progress has accelerated so much that sometimes people simply cannot keep up with the rapid changes" - nonsense. On the contrary - a stupor is coming, it is becoming more and more difficult to make semiconductor technologies smaller, Moore's law has died like a mammoth. Good morning. And the quantum theme is still a piece of crap.
    1. 0
      17 January 2025 16: 14
      Quote: Spicy Valenok
      On the contrary, a stupor is setting in, it is becoming more and more difficult to make semiconductor technologies smaller, Moore's law has died like a mammoth.

      This is good.
      Instead of training crappy models on "everything they collected from the Internet" while throwing money at the problem, you'll have to think (oh, the horror belay ).
      The same with programmers, instead of "buy another 32GB of RAM" you'll have to think "where did we screw up so badly".
      Well, in short, it’s about the same as in the 90s (and earlier). laughing
  5. 0
    17 January 2025 15: 59
    Without comparing, say, a top server processor with these qubits, they mean nothing.
    Moreover, judging by:
    The device is equipped with a sophisticated control and cooling system that maintains superconductivity at temperatures close to absolute zero (-273 °C).

    You also need to compare it with a car accelerated under liquid nitrogen (but stable, and it didn’t work and burned out). laughing

    I suspect that at best there is something like cryptography/fuzzy logic (and even then the machine will be bigger than the first supercomputer), and at worst - super-fast factorial calculation.
    It is not universal and complex. On transistors it is somehow more reliable and simpler (production\debugging\use).