Zvezdoschet: Russia is developing a quantum solver for business and industry

MIPT scientists and their partners are creating a quantum technology that will help make a breakthrough in logistics, energy and finance. Currently, research is being carried out on the basis of an already developed and implemented optimizer solution for the banking industry. Experts note that the technology will make it much faster to calculate tasks with tens or hundreds of millions of variables. For more information, see the Izvestia article.

What is a quantum optimizer?

Russian scientists are exploring technologies that will make it possible to create a quantum version of an existing solution — a universal optimizer, that is, a set of tools for solving specialized tasks. The technology itself is also called a solver or solver.

There were no own analogues of such a development in Russia before, the press service of the T1 AI development company noted. The country mainly used solutions from large foreign corporations (Western or Chinese, and in some cases also open source solutions), which were purchased under license. Therefore, the domestic universal optimizer will solve the problem of import substitution.

This optimizer will make it orders of magnitude faster to calculate tasks with tens or hundreds of millions of variables. Therefore, scientists are conducting various studies to identify problems to which quantum computing can be applied. The results of some studies show its significant advantage for solving a variety of tasks, while it is possible to single out logistics and transport, in which a set of tools and corresponding algorithms can significantly exceed existing methods, the developers said.

One of the main issues is to learn how to solve so—called NP-hard problems, that is, those whose solution by modern methods does not fit into any time frame. In particular, we are talking about decrypting messages, hacking information transmission systems, the developers said.

According to Alexey Chernov, project manager and Deputy head of the Laboratory of Numerical Methods of Applied Structural Optimization at the MIPT Phystech School of Applied Mathematics and Computer Science (FPMI), scientists are exploring the possibilities of practical implementation of quantum-inspired algorithms (computational methods based on the principles of quantum mechanics, which are used to solve problems on classical computers) optimization, as well as identifying areas their effective use.

—We hope that such research will make it possible to make the universal solver truly universal by adding a quantum component to it,— added Alexey Chernov.

Currently, the optimizer is already being used as a tool for solving specialized tasks, for example, to optimize the loading of production lines, inventory, staff work schedules, electricity generation, planning and managing the production of petroleum products, building financial portfolios, and similar goals. Experts assume that the expansion of the optimizer's algorithms will occur gradually, for specific industries and requests. At the same time, many existing algorithms will need to be improved for industrial use, the press service of the developer noted.

How is the creation of the optimizer progressing?

In five years, when quantum devices can become a reality, these developments may be in demand and will allow for a breakthrough in logistics, energy and finance, and will provide business and industry with a breakthrough by orders of magnitude. This will not only close the issue of import substitution in complex computing algorithms, but also give Russia a technological advantage, said Ruslan Gabbasov, CEO of KOR.

— We are studying how existing algorithms can be adapted to the solver architecture: we are testing hypotheses, testing approaches, and looking at how classical methods can be complemented by quantum principles. If these ideas can be implemented, it will become a new milestone towards the creation of domestic solutions based on quantum technologies, which will make it possible to advance optimization to another, higher level," said Sergey Golitsyn, head of T1 AI.

Andrey Raigorodsky, Director of the MIPT Faculty of Physics and Technology, Doctor of Physico-Mathematical Sciences, sees prospects for the development of quantum technologies in the field of AI.

— Our scientists are constantly playing for promotion, they will be the very visionaries who create trends. Our Phystech is about creating intelligent special forces that can become the creator of the agenda of the future. And now it is obvious that significant changes are also coming in the field of AI. For example, many specialties will die, like the driver's profession, and unmanned vehicles will take its place," the expert said.

Director of the ANO "Sports and Methodological Center "Department of Esports" (conducts comprehensive scientific and practical research in the field of high technologies, esports and AI) Victoria Beresneva believes that the new optimizer will be a strategic investment in the future: it will allow to form a scientific base, adapt algorithms to real quantum devices and prepare for the moment when such technologies will become an industrial standard.

As noted in the press service of the developer company, a quantum optimizer may appear in five years.