In the experimental building of Shanghai Research Institute of the University of Science and Technology of China, located in Pudong New Area, Shanghai, hundreds of components seem to grow randomly on a workbench of less than 3 square meters: some are bare and bare, while others are supported by wave plates like magnifying glasses, with different sizes and shapes, and the surrounding frame beams are covered with optical fibers. It’s hard to imagine that this is the main body of China Optical Quantum Computer, which has just come out, and has become a shining star in the scientific and technological circles of China and even the world.
On May 3rd, the developer of this computer — — The Institute of Quantum Information and Quantum Science and Technology Innovation of China Academy of Sciences announced here that Academician Pan Jianwei and colleagues Lu Chaoyang and Zhu Xiaobo of China University of Science and Technology, together with Wang Hao Hua Research Group of Zhejiang University, have built this quantum computer based on single photon, which is the first optical quantum computer in the world to surpass the early classical computers.
At one time, comments came one after another: "China scientists are once again at the forefront of innovation", "Quantum computer opens a new China speed" and "Quantum computing will completely change the future application prospects of mankind" & HELIP; … Even Pan Jianwei, the focus of this achievement, mentioned in an interview that "quantum computing research is like mushrooms after rain, and it has reached a critical moment of explosive development." So what can this "made in China" quantum computer do and what will it bring us? China Youth Daily Zhongqing Online reporter interviewed relevant experts.
Bicycles and airplanes
The reason why people are full of expectations for quantum computers is not unrelated to the problems encountered by traditional computers: in recent years, traditional computers have gradually encountered a series of problems such as power consumption bottlenecks and communication bottlenecks, and their performance growth is becoming more and more difficult. The demand for high-performance computing technology to explore new physical principles has emerged.
Quantum computer is a key for scientists to try to open infinite possibilities in the future. Theoretically, quantum computers can easily "kill" traditional classic computers: someone once made an analogy. If the speed of traditional computers is bicycles now, the speed of quantum computers is like airplanes.
The secret of "second kill" lies in the word "quantum".
The so-called quantum refers to the most basic unit of matter, which is inseparable. The well-known microscopic particles such as molecules, atoms, electrons and photons are all manifestations of quantum.
Pan Jianwei said that quantum has a characteristic that determines its ultra-fast computing power, and this characteristic is "quantum superposition". This is the fundamental difference between the quantum world and the classical world — — The famous "Schrodinger cat" vividly described this paradox. In the classic world, cats are either alive or dead, but a quantum cat can be in the superposition state of "dead" and "alive".
With this "nature" of superposition, quantum computing has the ability of parallelism, that is, synchronous computing can be realized. According to scientists, classical computers can only be "either-or" in binary algorithms: either 0 or 1. But quantum computers have the ability to "coexist".
For example, in order to "find" the correct information, the classic computer we commonly use at present needs to list and verify all the possibilities when extracting a problem to be solved, which is equivalent to a person with two hands who can only do one thing at a time; Quantum parallel computing can directly calculate and extract the corresponding information, which is equivalent to a thousand-handed Guanyin with two N hands, and can do what two N hands can do at the same time.
How fast the latter is depends on the exponential increase of n.
At the press conference on May 3rd, Pan Jianwei and the participants relived the classic story about chessboard wheat — —
In ancient India, the king was going to reward the inventor of chess and asked him what he wanted. Pointing to the chessboard in front of him, the inventor said to the king, Please put one grain of wheat in the first cell of the chessboard, two grains in the second cell, and four grains in the third cell, and then each cell will be twice as large as the previous one, and so on. When all the grains in the 64 cells on the chessboard are filled, you will give me a reward.
The king thought it was simple at first. Unexpectedly, when people brought bags of wheat to start counting, the king found that even bringing all the wheat from India could not meet the requirements of the prime minister — — If you count two pills a second, it will take hundreds of millions of years to count them all.
By the same token, Pan Jianwei said, with the increasing of N, the computing power of quantum computers will increase exponentially, and some large-scale computing problems that classical computers can’t solve will be "solved". He takes factorization of large numbers as an example. It takes 150,000 years for a classical computer to decompose a 300-bit large number, and it takes only one second for a trillion quantum computers to decompose this large number.
ideal and reality
According to Pan Jianwei’s description, quantum computers are also expected to play a big role in the large-scale computing problems of human society — —
In the field of public safety, quantum computing can instantly process 6 billion face pictures in the monitoring database and identify a person’s identity in real time; In the field of public transportation, quantum computing can quickly analyze and predict complex traffic conditions, so as to dispatch integrated transportation systems and avoid road congestion to the greatest extent, and so on.
However, these are only the "ideal" prospects for scientists at present, and quantum computers in reality are far from reaching this application level.
Even the China Optical Quantum Computer only beat the speed of the early electronic computer, which was 10 to 100 times faster than the first electron tube computer and the first transistor computer in human history, but it was "much slower" than the current smart phones and laptops.
In the early 1980s, American physicist Feynman put forward the concept of quantum computer, but after more than 30 years’ efforts, it is difficult for the best quantum computer in the world to complete some "seemingly childish" practical mathematical problems, so that people who are skeptical about it give this statement: Although quantum computing is a revolutionary technology, it always seems out of reach.
Now, things seem to have changed. At the beginning of this year, Nature magazine said in an article entitled "2017 will be the year when quantum computers come into reality from the laboratory" that quantum computing has long been considered as a technology that will be realized after 20 years. However, 2017 may be the year when this field changes its impression of "limited to research".
This change was confirmed in Shanghai, China on May 3rd, when the world’s first quantum computer surpassed the early classical computers came out.
Pan Jianwei said that there is a saying in the scientific community that there are "three steps" in quantum computing research as far as computing power is concerned. The first step is to surpass the computing power of the first electronic computer, the second step is to surpass the computing power of the commercial central processing unit, and the third step is to surpass the computing power of the supercomputer.
Now, Pan Jianwei team has finished the first step. He said, "At present, what we have achieved is only the first step, but this small step is an important step."
Commenting on this achievement, the reviewer of Nature Photonics said: "This quantum computer developed by scientists in China can be said to be the ENIAC in the field of quantum computing." The latter is the world’s first electron tube computer, which opened up the era of classical algorithmic computers.
The milestone significance of China, a quantum computer, can be seen.
Catch up and surpass
In fact, there are many schemes to realize quantum computing in laboratories all over the world, but the difficulty lies in the "manipulation" of microscopic quantum States. At present, Pan Jianwei’s team is working hard in three promising directions: photon quantum computing, ultracold atomic quantum computing and superconducting quantum computing.
The optical quantum computer unveiled on May 3rd is a quantum computing prototype that manipulates five particles, namely five optical qubits. According to quantum theory, the computing power of quantum computer will increase greatly with the increase of the number of particles that can be manipulated.
Lu Chaoyang, a professor at China University of Science and Technology, told reporters that the team’s next goal is to realize the manipulation of about 20 optical qubits by the end of this year, and is committed to the design, preparation and testing of 20 superconducting qubit samples.
At that time, the computing speed of quantum computer will be close to the best commercial central processing unit — — That is, the "second step" mentioned above, catching up with the speed of the current laptop. Lu Chaoyang said that by 2020, it is hoped that about 50 qubits can be manipulated, and then it is possible to achieve the "third step" breakthrough, that is, to surpass supercomputers.
Previously, John Preskill, a physicist at California Institute of Technology, mentioned a word called "quantum hegemony", to the effect that when a quantum computer has 5 to 20 qubits, it can accomplish what a top supercomputer can do, and once it has more than 49 qubits, the ability of a quantum computer will be beyond the reach of a supercomputer.
"This is called ‘ Quantum hegemony ’ The task, the motivation for scientists to design this problem at first is to show the ability of quantum computers and compete with classic computers to see who runs fast. " Lu Chaoyang said.
Pan Jianwei said that in the next 10 years, it is very likely that the scientific community will be able to manipulate 100 qubits. By then, its computing power may be billions of times faster than the most powerful supercomputer at present.
Further prediction, Pan Jianwei said that if more than 1,000 quanta can be achieved in the future, scientists may be able to study how consciousness is generated, and may be able to give birth to unimaginable artificial intelligence.
But at that time, ordinary people probably still didn’t need quantum computers. Pan Jianwei said that taking the existing mobile phone terminals as an example, the computing tasks that need to be completed can be sent to the cloud through the cloud computing platform, and the quantum computers in the background can complete them.
He also admitted that just as people who invented abacus in ancient times would not think of computers today, it is difficult for him to accurately predict the huge energy of quantum computers in the future, and everything can only be seen.
林永健
