Quantum computing:a bubble about to burst?
At this moment, at a large intersection in Illinois, the United States, vehicles are constantly interspersed, and few people who pass by will realize that the place they pass is the booming United States.Quantum Information Technology Research Office, also on Yunfei Road in Hefei High-tech Zone, China, it is difficult for the scientific and technological developers and travelers to realize that quantum information technology is sprouting near this trail.
Application of quantum information
Under the US interstate highway, in an entangled state Photons (particles moving at the speed of light) are forming a fiber optic cable, one of the longest land-based quantum networks in the United States, traveling back and forth to the U.S. Argonne National Laboratory in the next town.
At this moment, the researchers want to use this 83-kilometer-long quantum test base station (and other similar sites) to prove that you can encode information in a quantum state in one place (for example, in photons) , And then send it to other places and get this information in its entirety on the other end.
Of course, there are many difficulties in this move. Scientists need to overcome difficulties such as frozen soil, solar radiation, and vibration caused by vehicle traffic. Fortunately, once the researchers successfully confirm the above point of view, the new communication methods they bring will make a technology such as 5G a thing of the past.
At the same time, researchers in other laboratories are trying to introduce algorithms into qubits and convert them correctly when they complete their calculations. Once the move is truly successful, they will have a new type of computer-a quantum computer.
For many years, physicists have clearly known that quantum mechanics The principle can completely change the calculation method and the Internet, and the mature manipulation of qubits can greatly reduce the running time of the algorithm. Since any interference will destroy information, stable photons can quickly transmit information around the world without intrusion during the transmission process.
For the rest of us, the quantum revolution seems to have gone from a boring scientific theory to the sharpest frontier. But as everyone knows, we may even be experiencing some kind of quantum bubble-it may burst at any time.
Before 2017, most quantum tests were only performed through idle fiber optic cables. No one could make qubits, like classical computers, reliably process information.
The confusing advance
Now, there are dozens of quantum computers in the world. Run, several of which can be accessed by software developers through familiar servers, such as Amazon Web Services accounts.
In the past two years, the United States has invested more than US$1 billion (about 6.6 billion yuan) of government funds for quantum information research, and quantum computing startups have also completed multiple rounds of venture capital. And IBM announced that it is building a computer with more than 1 million qubits, and currently there are only 60 qubits at most. The Chinese government also announced a national-level quantum information technology policy plan in October this year.
Despite the rapid development, many people working in the emerging field of quantum information science say that the information processing technology of qubits is not yet sufficient to replace classical computers and the Internet, because it is not reliable enough, and people are concerned about it. Awareness is not enough.
Jokingly, most people think that instead of buying a mobile phone with a qubit, it is better to buy a mobile phone with an Apple bionic chip.
Applications like qubits and other elementary particles, in most people’s concepts, are always classified as belonging to the field of scientific research, and it is difficult to establish a relationship with themselves. But it is hidden, the future direction, and our objective experience has never been stable and reliable.
Qubits:Much better than 0 and 1
Computer composed of qubits It is actually a collection of circuits, just like a classic computer is composed of bits. The input value is processed through a series of logic gates in the circuit, and each logic gate modifies the value to produce an output.
If you are trying to solve a complex algorithm, for example, software application testing. To run on a classic computer, you need to string together multiple 0 and 1 bits. However, if you use qubits to run the algorithm, you only need one superposition qubit to replace all the classical bits.
In improving quantum computing and communication, the most difficult problem to solve is the vulnerability of quantum states. We can only protect the moving quantum particles from weather and road vibrations in the test, and the thousands of kilometers needed to replace the existing Internet is another matter. Similarly, even in a controlled laboratory environment, no one can figure out how to make qubits work reliably.
As IBM demonstrated with the 27-qubit Falcon processor earlier this year, they perform well in certain types of calculations and are mainly used for testing. For example, researchers can ask them questions with known solutions and then verify their answers. But so far, qubits are too fragile to operate reliably in larger groups, which has restricted them to the testing phase forever.
The IBM Quantum Computing Experiment Group stated that in theory, as the number of qubits increases, we can explore more diverse quantum circuits. But the reality is that the problem of”qubit loss” means that part of each of the existing quantum computers is dedicated to solving errors in calculations, rather than performing calculations by themselves.
In order to solve this problem and release the full potential of quantum computing, some researchers are studying the addition of error correction codes, which have been implemented in some classical computers.
Others are exploring other ways to apply quantum physics to computing, which do not involve gates and circuits. One possibility is to induce quantum particles to ignore disturbing background noise (such as vibrations, temperature changes, and stray electromagnetic fields). A team from the University of Chicago announced in August that they had successfully implemented this induction.
Quantum annealing is another promising technology that uses fluctuations in quantum states for calculations. Some commercial quantum computers of D-Wave use this method. But they are also plagued by errors, and so far can only effectively solve certain types of algorithms, such as finding the shortest path between a set of points.
Last year, Volkswagen used the D-Wave method in a trial in Lisbon, Portugal to help buses avoid traffic jams. This experiment was successful, although it was limited to bringing participants from the airport to the conference center.
Previously, the most notorious qubit loss appeared in the experiment nicknamed”sycamore” in October 2019. At that time, Google researchers announced that they had completed the benchmark test of a 53-qubit quantum computer in 200 seconds. This test requires a classic supercomputer to take a few days to 10,000 years.
So Google claims to have achieved a quantum advantage. Quantum computers can run algorithms faster than classical computers without making any mistakes. This is a milestone in the field of quantum information science, which Google CEO calls the”hello world” moment of quantum computing.
However, soon after, researchers questioned this experiment, which caused controversy in the industry.
Massachusetts Institute of Technology physicist Wiliam Oliver said that quantum advantage does not lie in whether it exists, but when it collapses. Most people think Google did it, but if they added a few qubits, it would be impossible to do it. He also believes that the benefits of quantum computing are not just beyond classical computers. The real milestone should be that quantum computing can run arbitrary algorithms without errors at any time.
Venture investment and government funds
If there is quantum The bubble is not only due to the emergence of sycamore-style academic research, but also private companies that are simultaneously promoting the development of real-world quantum applications (such as avoiding traffic jams), which intensifies the expansion of the bubble.
At least since the 1980s, when Argonne physicist Paul Benioff described the first quantum mechanical model of a computer, we knew the advantages of quantum mechanics in computing. However, this technology seems to have just the right appeal for small startups and large enterprise groups.
William Hurley, a software systems analyst who works for both Apple and IBM, founded Strangeworks in 2018 as a community center for developers who study quantum algorithms. He personally believes that now is the most exciting moment in the quantum age.
He also said that more than 10,000 developers have signed up to submit their algorithms and collaborate with others. Among them, in order to test their quantum algorithms, the startup Rigetti provides users of Amazon Web Services with access to one of their computers. The service, called Amazon Braket, debuted in August and its customers include Volkswagen and Fidelity Investment Group.
Quantum information technology is so attractive that large companies are dividing up entire research departments as a way to stay competitive. Researchers at JP Morgan Chase have developed quantum algorithms for every part of its business, from encryption to security for option transactions.
JPMorgan Chase officially stated that it is currently in research mode and hopes to be prepared when quantum advantage comes. The company is optimistic that quantum computing can improve option trading, and in this financial sector, speed and accuracy are crucial.
All these activities have received financial support and incentives from U.S. and overseas taxpayers. The U.S. National Quantum Plan was established in 2018 and its scope is wide (it calls for the development of a”ten Annual Plan” to accelerate the development of quantum information science and technology applications) and generous ($1 billion has been approved so far).
The military has also provided a lot of support. For example, the US Department of Defense Advanced Research Projects Agency (DARPA) has allocated nearly 20 million US dollars (about 130 million yuan) so far this year to promote the development of quantum computers.
Rigetti claims to own the only factory in the United States that specializes in the production of quantum integrated circuits, which has attracted government funds and venture capital. In March of this year, the company received US$9 million (approximately RMB 60 million) from DARPA, and subsequently completed a US$79 million Series C financing (approximately RMB 520 million). And in August announced its plan to build a second quantum computer in the United Kingdom, which also received 10 million pounds (approximately RMB 86.2 million) funding from the British government.
The government and the military are particularly interested in building the quantum Internet. The Deputy Minister of Scientific Affairs of the Department of Energy issued a statement in July that they now have a blueprint to make the quantum Internet a reality. Eventually, the department plans to establish quantum test loops in all 17 national laboratories and connect them to create a basic nationwide quantum communication network.
However, quantum uncertainty represents a gamble, it is not yet ready to solve real-world problems.
In fact, if qubit loss is the nemesis of quantum physicists, then access challenges are the bane of corporate researchers. Whenever a quantum computer completes its algorithm, it needs to rest, otherwise quantum entanglement will completely collapse. The superposition state of qubits will disappear. The process of disappearing coherence between superposition states in the classical world is called quantum decoherence, which further proves the fragility of quantum computers.
In September this year, physicist Wiliam Oliver of the Massachusetts Institute of Technology and other scientists jointly announced that they believe that harmless radiation from ordinary objects such as concrete walls accelerates this decoherence. It is impractical to transfer a quantum computer into a radiation-free bunker, and research on other potential remedies (such as background noise deception) has just begun.
Therefore, in the foreseeable future, quantum computers will have to be reset frequently. If you want to provide quantum computing cloud services like Rigetti and Amazon, that means your customers have to wait a long time.
The fact is that a method is needed to transfer algorithms from classical computers to quantum computers, which exacerbates the access challenge. Initially, Rigetti allowed customers to submit a line on the regular Internet and then get the results. However, what is really needed is a tight loop that occurs between the classic computer used to submit the results and the circuit that runs again, and putting the public Internet in it is a huge deterrent.
Rigetti officially stated that thanks to recent improvements, Amazon customers will be able to avoid some lag time, which allows Rigetti to evaluate thousands of circuits at the same time and return the results to the customer’s classic in milliseconds. On the computer.
But they admit that unless you completely transition to a quantum system, the only way to completely avoid lag time is to integrate quantum and classical systems, and it will take decades to complete this achievement.
Can quantum computing go out of the laboratory?
In contrast to the rapid development of quantum technology in the past three years, the next technological leap is a long one. But progress is relative. More than a century has passed since the first electronic circuit powered by a vacuum tube in the 20th century to the semiconductor manufacturing technology that used a stamp-sized A12 bionic chip to power the iPhone. Therefore, do not prematurely define the future state of affairs.
David Aschwalom, a physicist at the University of Chicago and the leader of the Argonne Quantum Test Loop project, is also the main author of the aforementioned background noise deception technique. He pointed out that the current state of quantum information research is equivalent to a classic computer with only a few dozen transistors in the 1950s (modern notebook computers have billions of transistors).
But he pointed out that equivalent quantum machines with dozens of qubits are expanding on a”highly nonlinear” scale. When someone invents a quantum computer with about 200 qubits, We can run algorithms in more states than there are atoms in the universe.
IBM released a roadmap in September that shows how it will upgrade from 27 qubit computers this year to a 1121 qubit processor called Condor by 2023. . Ultimately, IBM hopes to build a completely fault-tolerant computer composed of 1 million qubits. This is also a difficult project for IBM’s own engineers.
In addition to making so many qubits work well together and maintain consistency, Condor also needs to improve the supporting system and physical architecture. A quantum computer with low-temperature superconducting qubits will be a behemoth, with numerous dilution bridges and cryogenic cooling chambers, and hoses connecting all the parts together.
All computers generate heat, but quantum computers are veritable melting pots-first of all, they are extremely susceptible to radiation and temperature fluctuations. IBM pointed out that today’s commercial refrigerators cannot effectively cool and isolate a million-qubit computer. Therefore, the new refrigerator is also on the quantum road map, which may be a super refrigerator nearly 3 meters high and 2 meters wide.
The future vision of quantum power
This research is very real Yes, but it is unimaginable for most of us. In order to verify their algorithms, 10,000 software developers who share algorithms on the Strangeworks platform need to find the free time of the quantum computer and verify the accuracy after running. Or they can choose to rewrite these algorithms and let them run on classic computers.
Even if quantum computers succeed in completing these tasks faster than classical computers, in the long run, this superiority of quantum computers is not important. As for the quantum Internet, the discarded fiber optic cables of the test platform remind us that challenges still exist.
Speaking nicely, the current state of the quantum field is mature enough to accept setbacks. To put it harder, this is a bubble that may disappoint taxpayers and venture capitalists alike.
But from a consumer’s point of view, quantum technology is different from 3D TVs or virtual reality devices. It doesn’t require you to buy a new TV or use your entire room in your house to play video games. Even if it turns out that it is impossible to use quantum processors to make mobile phones or use photons to reliably transmit information, the public can still feel the contribution of quantum to daily life.
Rigetti believes that ordinary consumers may view quantum computing in their lives in one of two ways.
The first is to find the shortest path between a set of points. Whether it is sending online ride-hailing drivers to communities with high demand, or driving buses on blocked roads, improving the ability to reach multiple locations as quickly as possible is a great boon to modern life. The second is the improvement of consumer products (especially medicines). Quantum computers are good at building molecular models for drug development.
These two situations and other similar situations show that in the future, quantum physics will not replace the information technology infrastructure we have today. In fact, the future of quantum is the future of data centers, not what mobile phones or laptops will look like 50 years from now.
In other words, perhaps there is no quantum bubble about to burst. Today, everyone can understand the importance of supercomputers, although it is not easy to accurately describe their role. Regardless of whether we have achieved quantum advantage or invented indestructible photons, there is no doubt that quantum technology will improve such machines.
In the past, it might take several months to get to another city. The invention of the train shortened the time drastically. Now, you can compare a train to a classic computer. No matter how far the train can go, you will eventually hit the ocean, so you need to take a plane.
This is quantum computing:a revolutionary technology that allows us to cross the ocean of metaphors. Even when it reaches its maturity stage, it may still rely on classic computers to perform daily tasks, just like when we take a flight to a foreign country on vacation, but on the way to local supermarkets and shopping malls, we still choose to drive vehicles.
If the recent investment in quantum computing research makes you want to buy a quantum-driven iPhone, then you may be living in a bubble. The quantum breakthroughs of the past few years have shown that in the not too distant future, although changes will not be obvious, they are more revolutionary. A new vaccine may be born in a few days, not a few years; stock trading may happen in an instant… This makes it worthwhile to spend billions of dollars today to accelerate photons and build giant refrigerators.
Follow quantum guest to obtain first-hand information in the quantum field p>