Dinghong introduces that quantum algorithm can make full use of the parallel advantage of quantum computation, and greatly improve the probability of the result of quantum computation.
The way to realize quantum computation is physical hardware, and the relative advantage of superconductivity is good scalability and long coherence time. The advantage of topological quantum computation is that the theoretical coherence time is long and the error rate is low. But the disadvantage of topological quantum computation is that the experiment is in the initial stage.
Referring to the application of quantum computing, Dinghong says quantum computing can improve machine learning and artificial intelligence, accurately predict weather, accelerate space exploration and solve traffic jams. (Tan night Cold)
The following is the full text of the speech:
Dinghong: Very pleased to be able to come to the future Forum annual meeting, to do a report on quantum computing, in fact, I do quantum computation is just getting started, I have done a lot of more is to do superconductivity, in recent years, did many topologies, now want to combine the topology and superconductivity is the topological superconductor, A topological superconductor can be used on a quantum computer, so now I'm going to do quantum computing, and my topic is called quantum computing, the engine of the next industrial revolution.
I'm from the Physics Institute of CAs. We all know that in the past there have been several times the industrial revolution, accurately said that there have been three times, the first is the steam engine as the main power of the factory production era, the second is the generation of motor and internal combustion engine, mankind has entered a large increase in productivity of the electrical era. The third time is the information age that we are very familiar with, a third industrial revolution, dominated by computers, has now evolved further, with the advent of the fourth industrial Revolution, which began with the Internet, big data and artificial intelligence, and the third and fourth times, the computer has played a leading role, but this year, computer development, Now I'm going to run into a bottleneck. We are familiar with the Moore's Law, that is, every 18 months, the number of components on the integrated circuit can be increased by one fold, the computational performance is increased by one times, which makes the size of the components on the computer more and more small, now has entered the nanometer era, has been to less than 10 nm era This brings up two questions, the first of which is that the classic computer is the unavoidable heat energy problem, I'll talk about it in a minute. The so-called Wave Delphi Machinery, the second is with the effect of the components have quantum effects, electrons on the small scale will occur quantum wear phenomenon, so that a classic bit will lead to the eventual failure of Moore's law , this time someone proposed a quantum computer, why do quantum computer, because in the classical computer has an irreversible, is unavoidable heat loss, for example do this operation 0 and 1 multiplication answer is 0, 0 and 0 multiplication, also 0, upside down say I know the result of this behind 0 is a variety of multiplication, 1 and 0 or 0 and 0, which is irreversible, wave de principle said that if there is the loss of this bit information, is irreversible process, the loss of information, there must be a part of the heat in the form of emissions, so that there is a hot limit problem, but the quantum computer is completely reversible in principle, Can transcend the Derges limit of this wave.
In addition, for the quantum world, it is a very complex, in fact, the famous physicist Nobel laureate of the 80, who proposed that the calculator to pursue quantum computing is the best way to simulate the quantum world, which is the world of the original concept of quantum computing, and we look at what is quantum. The so-called quantum is the more familiar photons, electrons, atoms constitute the basic unit of matter, but also the basic carrier of energy, the quantum world has two unique landscapes, a large is called quantum superposition, here is a more popular example, that is, Schrodinger, life and death that is determined, quantum inside if there is (quantum cat sound ), that is, life and death will have mixed states, that is, the fate of unknown. The second one is quantum entanglement, what does that mean? Two quantum entangled, this entangled state of the interrelationship, one is the self hangs upwards, a self hangs downward, this schrodinger cat's particle, this cat entangled one is dead, the other cat becomes alive. The quantum of becoming entangled team is too superimposed, is uncertain, can be alive, can also be dead, once you find that it is a live cat, another cat is dead cat, this can be divided infinitely far, very far. It seems that quantum entanglement is very hard to imagine. Some people say that it is possible to use the concept of quantum motion to understand the fact that space is far away, but they have a way to move through a moving, actually very close.
For a classic bit, 0 and one are identified as a set of bits, a group of bits can represent a state, when we do calculations, whether it is the calculator, the input and output of 88.8, are OK. The same is true of the computer, enter a certain number, output a certain number, but for a qubit, he does have uncertainty, except for 0 and 1, for example, this is where the self-study and downward is to represent 0 and 1, and one that says he can in this direction, in any direction of the ball, This arbitrary direction can be projected out, that is, 0 and 1 of the linear combination, which can be 0 and 1 at the same time, because any direction is the direction of 0 states, there is a part of 1 states, 0 and 1 states at the same time, this will lead to the quantum computer exponential growth, This is a metaphor to say that India's chess in a chess word put a small board, the other one put two, four, eight, so that the 64 pieces are put out, how much wood? 184.4 billion grains of wheat, are very amazing, quantum bits have the same function, a qubit that stores two of information, two stores four, four stores eight, 16, and then 60, which is also a staggering number of people who put this 50 bit, because it can achieve 100 trillion times the classical calculation, If you can make full use of this quantum bit to do your 50 of qubits, in principle, you can do 100 trillion classical operations, so there is a very large advantage, the quantum algorithm also ensures that the parallel advantages of quantum computing can be fully played, And can greatly increase the probability of the results required for quantum computation, here is also a simple talk about two quantum algorithm, a called unordered database search, a large number of sorting, respectively, in 1996 and 1994, this is said to be the search for the database, the classic words, If we're going to find an unordered number, 12345, for example, put one or two is not, put the third, said is, and then find, so search down, the average 50 times, quantum search words, 1234 are put in the words, out of the words will have no, no, yes, of course, will do the iteration, Will prove to be an N, 100 of people 10 times can be searched. Here is the curve to show that Germany, this is what use, can be used as map navigation to improve efficiency, if the map navigation words have n road, the path is possible, that is n factorial, or 2 of the N-time, if used in this search, it is relatively fast. Because many, can open the square root to calculate. The second one is to improve the search speed of large data. This algorithm is mainly used to factorization, we all know factorization, for example, here to cite an example, 57 decomposition into 3x19, for green, so large a number, factorization is very difficult, do multiplication is very easy, but the product to make the factorization is very difficult to divide into two of the array is very difficult, Even supercomputers cannot be made, but for the proposed quantum decomposition, can, I have a quantum computer, I quickly decompose this large number into two, this decomposition speed is the red line, with this blue wire ratio, you see the Blue line is an exponential form, the above is 10 of the 35 times, red lines stay still very slow, So you ask what is the use of this factorization, the factorization has a very large use, here is a RSA encryption, the most important is the use of factoring, there is a public key and key, the public key is the two-key index multiplication, give you the public key, you are not the decomposition of these two keys, I tell you the key. So for a computer like this, is very difficult, for example, the study of the general 1024, in 2006 with the classic computer to break, is 10,000 years, is now relatively short, 38, 24 can be deciphered, now said it is best not to use this, because there is someone to decipher, later can be developed to the number of days, for quantum computing For a quantum computer, it can be deciphered in 10 minutes, because you say I'm going to raise this number to 1000, 2000, 4,000, and over the age of the universe. It's almost impossible, but you see, for a quantum computer, there's no need to develop it faster, from fractions to hours. In this way, it is very easy to decipher RSA's password, that is, to make this asymmetric factorization symmetric, so that RSA's key is equal to the public key, which is an important role.
Quantum computer utilizes quantum coherence phenomenon, you use is decoherence, this cat is dead or alive, if you look at it and make sure that this is OK, the classic is not that the output and injection must be a definite value, either 0 or 1, but quantum words, you have to give him a decoherence, it might say a mixed state So this 0 and 1 is 50%, you test is 5 times 0, 5 times is 1, then you say half live, one half dead, 50% of the cat alive, 50% of the cat died, so in this case, because there is this feature, can also be said to have this shortcoming, so the quantum computer does not need a large number of parallel budget middle, It actually has no advantage compared to the classic computer, we see a video, do not need to use quantum computation, do not need to calculate in parallel. So quantum computers are fundamentally incapable of replacing classical computers completely. The two applications are different, complementary, a little similar to the incandescent and laser, incandescent and laser is different, incandescent is an irrelevant light emission, laser is coherent light, and computer and quantum computer relations similar.
But I can't say that I use lasers instead of incandescent bulbs, which is no use, but the incandescent lamp can do a lot of things, lasers and lasers can bring great technological revolution.
Of course, quantum decoherence also brings about noise problems, this is one of the biggest problems in quantum computing, that is to say that its environment will bring it and how to solve the problem, one way is quantum, such as quantum error correction and quantum avoidance, which makes quantum errors difficult to occur in topological quantum computations. There are now five ways to achieve quantum computing, one of the main directions called superconducting quantum computation, a more forward-looking, is topological quantum computation, but the topological quantum computation at one of the most cutting-edge, has not realized quantum bit, superconducting quantum computer is now the mainstream of the industry, has developed rapidly in recent years, starting from 2012, This Google and UCSB cooperation, 2012 from four bits to 2014 5 bits, 2015 9 bits, China last year and this year achieved 10 superconducting bits of quantum computer, now has IBM is 16, 17, to Intel also said he can do 17, at least according to Google, He said they were expected to make 49 superconducting bits by the end of the year. I said that these 50 100 trillion times, so-called quantum hegemony, is not able to do so, although the number looks very slow, but I do not know, as this number increases, it is actually behind an exponential increase, to 50 will be very. Because of this, in recent years, the business community has been very much involved, Google, Intel, IBM, and of course, also saw Alibaba, in fact, I am behind the speaker is Alibaba's chief quantum scientist and Dayang cooperation, July 2015 jointly established the Chinese Academy of Sciences, Alibaba quantum Computing Laboratory, There is also a lot of input from governments, China also has a lot of input, 1.9 billion, in 2013-2015, there will be more, not only the quantum satellite, is now preparing for quantum information science, the National Laboratory, headquartered in Hefei, which is their design, this is the quantum of photon symbols.
Finally, I have two minutes to speak briefly about topological quantum computing. The concept of this topology is just as the cup can be continuously changed into a doughnut, similar to the topology, in which there is a circle, some people have been studying in a few years ago, that is, this circle, the winding, this knot, is a periodic table, you see here are coded, there are 0, three, four, five, this five there are different species. This was done to explain the periodic table of the elements, which, of course, was ultimately not related to the element cycle. But in fact, later thought to be useful, there is a use of the so-called hyper-suspension theory in high-energy physics has been some inspiration. Another use for topological quantum, that is, the knot as a topological qubit, how to do? This is what can be done with two dimensions, here's an example, Einstein and his students, Einstein ran around, on the map can run like this, but we know that Einstein said time and space is the same, there is a timeline in this two dimensions constitute a three-dimensional space between time, you see how this axis is going, Is this way, the student does not move to make this white line, so that the words will form a knot like this, but they both do not move, the light looks at the initial state and the last state, two identical, but the time period is different, resulting in different topological states, which is two of the quantum states of unequal valence, can be said to be a topological quantum bit state, the second state, can do, In this way it becomes a so-called arbitrary value of the imagination, what's the benefit of this, the process of compiling or the process of saying this knot, just about the topology, and its specific shape, you say there's some noise, and there are some disturbances that won't change the result of it, so this will receive a quantum computation of topological protection, There is a natural high fault-tolerant phenomenon, with what quasi particles, it is necessary to use the Ma Yue rana quasi-particles, then Ma Yue Rana through an equation confirmed that the ordinary in a mirror, positron and negative electrons, the nuclear opposite, for Ma Yue Rana, mirror the same, this is also named the Angel particle, the same as the picture, How to achieve in the material, of course now there are three methods, one is called the true topological superconductor, the second is to use the space static effect of the topology and superconducting combination, 2008 to achieve Fu Liang and his teacher, This year we officially have a third method is to use the guidance space of the neighboring effect to achieve there are three ways, some of the recent articles we have done have gradually proved that the presence of the equine nanoparticles in iron-based superconductors is a process of several experiments, and I am not speaking because my time has come, and finally, the application of quantum computing, Whether it is artificial intelligence, or machine learning, accurate prediction of the weather speed up space exploration to fast algorithm, solve the traffic jam problem, here will bring great changes, so is the next industrial revolution an engine, thank you.