For a few months, quantum computing has been the subject of hype. The first quantum computers are there, and their creators, IBM, and Google in mind are quick to remind us. But are we ready?
Why are we interested in quantum computing?
The theoretical bases of quantum computing were introduced in the 1970s. At the time, and for twenty years already, the first steps of computing were based on classical physics and on the mathematical theory of computer science. information. Today's computer science is therefore based on these concepts, which are starting to date ...
Quantum physics is based on a break in the binary model - the famous bits of the binary code, 0 and 1 - challenged by the exploitation of the particular behavior of objects at the atomic scale. Concretely, in the quantum dimension, a particle can exist in many states at once, called "superimposed". Two particles can also be 'entangled', and under these conditions be simultaneously affected by a change of state.
From this state of matter, researchers deduce that information stored in quantum bits, or Qubits can be exploited to perform calculations on an exponential number of states. Quantum phenomena can thus realize in near real-time complex calculations, even impossible in the classical state of computing.
Quantum computer technology is here
For this, the quantum computer must obviously be able to create and manipulate Qubits. However, to design a quantum computer, it was first necessary to have technologies capable of supporting quantum theories. These technologies have come with the evolution of R & D., In particular, the production of superconducting materials to compose circuits in which the two distinct electromagnetic energy states which constitute a Qubits coexist. This has been made possible by phenomena such as the simultaneous circulation of current in a clockwise and counterclockwise direction.
Today, in the labs, the first quantum computers support 50 Qubits. By comparison, this is about the state where a conventional supercomputer would turn in about 50 bits. And it's mostly, in theory, the point where it becomes almost impossible for a 'normal' computer to solve problems.
The promise of quantum computers is to perform calculations beyond those of conventional supercomputers. Many domains are waiting for these computational capabilities. For example, to simulate the behavior of matter to the atomic level to discover new materials. Or to break the security and cryptography codes (which is why some US governments fund quantum research!). Or even more effectively analyze the data in Artificial Intelligence.
Quantum supremacy' is not for tomorrow ... still too many mistakes
However, the quantum still has to solve big difficulties, even limitations. First of all, it is very difficult to maintain Qubits in time. They tend to 'decode', ie to lose their quantum nature. Moreover, the 50 Qubits must work perfectly and together, which is not yet the case, quantum computers are indeed subject to errors that must be corrected. In the opinion of many experts, quantum computing promises to be able to reach the exponential, but it also carries within it the exponential means that it goes wrong!
In addition, the quantum computer carries a paradox that makes us doubt its usefulness: for many calculations, the quantum would be slower than the classic! This is bad news, the quantum does not speed up all the tasks! Thus classical simulations on a computer also conventional turn out to be complex to adapt to quantum computing, and do not offer the expected controllable precision.
As in the field of R & D, many chemical reactions and material properties are determined by the interactions between atoms and molecules, the quantum computer should be imposed since its model is itself governed by quantum phenomena. Eventually, it should allow modeling classical models as quantum. And therefore prevail in most areas that involve heavy and complex calculations.
The quantum nightmare
For the moment, we are far away, because quantum computers require not only different programming languages but also a fundamentally different way of thinking about programming. Isaac Chuang, a professor at MIT, was recently noted by saying on the quantum computer: " It's no longer a physicist's dream, it's an engineer's nightmare ".
That's why, even though IBM, Google, and others are bringing their quantum newborns online, we can not do much today. Ambitious dreams, even if ultimately realistic, from quantum computing to advanced AI will have to remain in the dream state for now. Until a generation of students trained in this revolutionary technology can join companies to enable them to manipulate Qubits. It will come, but certainly not for tomorrow. In the meantime, it will be enough to dream about the prowess of R & D in the quantum sphere, while remaining mindful of the margin of error that technology will continue to support ... for how much longer?
Also published on Medium.