Quantum physics governs everything that surrounds us. Is it possible to turn the whole Universe into a quantum computer, will aliens notice it, and why such machines are needed at all - Jacob Biamonte, a Skoltech professor, one of the leading experts in this field, answers these questions and tells how he ended up in Russia.
Bright future
“I first came to Russia more than ten years ago, and not at all to do physics. I am fond of martial arts, including sambo, and came here to study and exchange experiences. Later I learned that there are all the conditions here to engage in advanced science, attracting scientists from all over the world to cooperation,”says the scientist.
Today he heads the Deep quantum labs, created two years ago within the framework of Skoltech to unite the efforts of Russian and foreign physicists, mathematicians, programmers and engineers studying the problems associated with the development of quantum computing systems.
“We do not deal with practice, but all theoretical and“software”aspects of quantum computing, and we interact with experimenters, including Skoltech scientists and specialists from Moscow State University, RCC and ITMO. We are open for cooperation and are ready to help any experimenters studying such issues,”the professor continues.
What is a quantum computer? By its very nature, it is radically different from classical computing devices, which allow simple or complex mathematical operations on numbers or datasets expressed as zeros and ones.
In the quantum cousins of classical computers, the principles of which were formulated more than 30 years ago by the Soviet physicist Yuri Manin, information is encoded in a fundamentally different way. Elementary memory cells, the so-called qubits, can contain not either zero or one, but a whole spectrum of values in the interval between them.
Promotional video:
As a result, the power of such computers grows exponentially: the behavior of a quantum processor with several tens of qubits cannot be calculated even with the help of the most powerful classical supercomputers.
For a long time, such machines remained the subject of science fiction and theoretical research by physicists, but in the last 15 years, scientists have made a breakthrough in creating qubits and in combining them into more complex systems. The most advanced versions of quantum computers developed at Google, IBM, and at Harvard University by Mikhail Lukin's group contain 20 to 50 qubits.
Timur Sabirov (Skoltech). Jacob Biamonte, professor of physics at the Skolkovo Institute of Science and Technology.
Despite these advances, the developers of these machines assume that full-fledged computing systems capable of solving any problem will not appear soon, in 10-20 years. Interestingly, this estimate has not changed since the late 1990s, but some new problems constantly arise, each time pushing aside the never-coming “bright quantum future”.
As Biamonte noted in his popular science lectures, he takes a special position: in his opinion, "useful" quantum computing systems will appear much earlier, but they will not be at all what the general public and the media imagine them.
“Today there is one big problem in physics, which is at the same time its main advantage. Experimenters run everything. For some reason, they think that experiments are more important to science than theory. Thanks to the money invested in this area, theoretical physics has been virtually destroyed,”says Biamonte.
The professor himself refers to himself as a representative of classical theoretical physics, whose ideas dominated science a century ago, at the first stages of the birth of quantum mechanics and modern Einstein's physics. In recent decades, people like him had to move to mathematics departments, where they are much more comfortable.
“Experimenters, including the creators of quantum computers, care only about their own designs. With a few exceptions, they are not interested in what is known about the ability of such devices in general. This affects their mentality and makes them give not rational, but emotional assessments,”the researcher explains.
For example, there is still no single clear evidence that quantum computers can outperform their classical counterparts in computing speed. At the same time, Biamonte specifies, if we generalize all simplified models demonstrating some aspects of this superiority, we will get quite convincing evidence in favor of the superiority of quantum calculators.
“On the one hand, Aleksey Ustinov, Aleksandr Zagoskin and other leaders in this field are right: a quantum computer is really not coming soon. On the other hand, in this case we are talking about universal machines capable of correcting their own mistakes,”the physicist notes.
The lack of such an ability in a computer, Biamonte emphasizes, does not make it absolutely useless or inferior.
Atomic adding machine
“There are countless examples of various quantum systems in nature that do not have this ability. Their behavior is very difficult to calculate using ordinary computers. Therefore, the creation of a quantum system that simulates such processes will allow us to carry out the appropriate calculations and get something useful,”says the scientist.
This idea is far from new - it was voiced by the famous American physicist Richard Feynman just two years after the publication of Manin's first articles. As Biamonte noted, experimenters have been actively developing such systems in the past few years, and theorists are thinking about where they can be applied.
Such analog computing devices, the so-called adiabatic computers, or "annealing" in the jargon of physicists, do not have to use quantum effects - for many problems, the classical interactions between atoms are sufficient.
“There are three types of computers of this kind - classical annealing machines, their quantum-accelerated counterparts, and full-fledged quantum processors based on quantum logic gates. The latter were created in IBM laboratories, the first - in Fujitsu, the second - in D-Wave,”says the scientist.
Biamonte and his Skoltech colleagues are most interested in third-class machines. Such devices, he said, are quite difficult to create, but they can be used to solve the most complex optimization problems: from machine learning to the development of new drugs.
“These machines are very interesting, but the first real devices of this type will only appear in a few years. On the other hand, it is possible to create classical and quantum annealers right now. And now, in practice, they remain the most useful of quantum computers,”adds Biamonte.
Many processes in particle physics, the researcher continues, are programmed by nature so that they optimize themselves, striving to reach an energy minimum. Accordingly, if we learn to control these processes, we can make a set of atoms or some other objects do these calculations for us.
“Why waste a huge amount of CPU time on such an optimization when a classical annealing device or a quantum device similar to D-Wave can do it? Figuratively speaking, why, when studying the wind, use a virtual wind tunnel, if we already have a real one? Many Russian companies are thinking about this, and we are actively cooperating with them,”the scientist emphasizes.
Successful completion of these experiments will pave the way for the development of quantum annealing agents, in which the principles of quantum physics are used to accelerate interactions between atoms and other particles. Of course, some scientific tasks will not be available to them, but they will be able to solve many everyday problems, such as traffic optimization or stock portfolio management.
Most observers, the Skoltech professor notes, believe that Google will win in the quantum race. Biamonte disagrees with this: representatives of the Californian company are very fond of talking about their successes, but they almost do not publish scientific articles and do not reveal the secrets of the device of their quantum machines.
In his opinion, IBM engineers are closest to the goal - the computers of this company really work, and they can be checked at any time through special cloud systems. But the scale is still quite limited, and these machines cannot yet be used to solve complex problems.
Thinking galaxies
If such "serious" systems are created in the near future, a logical question arises: what can they be made of, what size can they reach, and how will they affect our life?
According to Biamonte himself, there are no fundamental physical limitations for quantum computers (or annealing devices) with millions of qubits. On the other hand, it is completely incomprehensible how many qubits there will be in reality, since we are now in the very early stages of the development of quantum technologies.
“So far, we are trying to adapt the technologies already available in the electronics industry to work with quantum computers. However, no one is sure that this is the right way. There are systems that are much better suited for building quantum machines. They are, however, much more difficult to manage,”the scientist explains.
For example, special defects inside diamonds are almost as well isolated from the outside world as single atoms in the vacuum of space. How many such points can fit in one diamond and how close they can be to each other without interfering with the work of neighbors is still unclear. The answer to these questions determines whether diamonds will be used in quantum computers.
Really large quantum machines, as noted by the Skoltech professor, will solve not only practical problems related to everyday human life, but also the most interesting scientific puzzles.
Perhaps they will reveal the quantum nature of gravity and test Biamonte's theories of time symmetry by observing whether they are particularly disturbed when trying to break this symmetry or reversing time when performing calculations on such machines.
When humanity has coped with these tasks, what will science do next? This question, says Biamonte, is paradoxically related to the search for extraterrestrial life and how representatives of alien civilizations can signal their existence.
Imur Sabirov (Skoltech). Jacob Biamonte and his colleagues at the Deep quantum labs.
“Imagine that we will subdue all the energy and power of the universe. What will we do first? Of course, we can destroy ourselves, but there is a more interesting scenario. For example, we will have the opportunity to accelerate the Earth's motion to ultra-high speeds and leave a computer in orbit,”says the physicist.
According to the theory of relativity, time on the planet will slow down. If we spend tens of years in this state, a quantum computing machine or an ordinary computer in the “outside world” will work for several millennia. Moreover, this is not necessarily a man-made computer, its role can be played by various space objects - giant clouds of gas, for example.
“How often can you do this? There is no explicit limit to such a "speedup of computations", but we all know that the late Universe will not be a very interesting place for us. The stars will gradually start to fade away, and the galaxies will become invisible to each other due to the expansion of the universe,”the professor notes.
Similar reflections raise a natural question: if humanity can do it, what prevents aliens from doing the same? Accordingly, some traces of such "space" quantum computing or their classical counterparts must be present in space. What would indicate this, the giant quantum computers of aliens?
“I cannot give an exact answer to the question of what it might be or to suggest how to look for them. At the same time, the existence of such “universal calculators” seems to me much more probable than the spontaneous emergence of “intelligent planets” and other cosmic objects capable of being aware of themselves, which is often discussed by “quantum” philosophers,”concludes Biamonte.