Predicting the behavior of interacting quantum particles represents one of the greatest challenges of modern science, especially due to their complexity. Quantum many-body problems, which involve a large number of particles, are crucial for the development of new technologies such as quantum computers, but also for advancements in materials science and chemistry. Researchers at the Swiss EPFL have developed a method that not only allows for a better understanding of these systems but also provides a tool for comparing different algorithms that attempt to solve these problems.

Particularly challenging is finding the ground state of a quantum system, which is the state of lowest energy in which a system can exist. This state is crucial as it reveals which materials will be stable and indicates the possibility of the emergence of exotic phases of materials, such as superconductivity. Quantum algorithms such as Monte Carlo simulations and tensor networks attempt to approximate the solutions to these problems, but so far there has been no universal method for comparing their accuracy.

Revolutionary method: V-score

A new tool called "V-score", developed by a team led by Giuseppe Carleo, allows for a consistent comparison of the efficiency of different quantum methods. The V-score is designed to measure the accuracy of algorithms in predicting the ground state of a quantum system, combining information about energy and fluctuations within the system. Lower energy and fewer fluctuations indicate more precise solutions, and the result of the V-score facilitates the ranking of methods according to their accuracy.

As part of the research, scientists collected the most comprehensive dataset on quantum many-body problems to date. Simulations were conducted across a wide spectrum of quantum systems, from simple chains of particles to complex, multidimensional systems like frustrated quantum lattices, which are known for their difficulty in simulation. This benchmark not only showed which algorithms perform best for specific problems, but also highlighted where quantum computing can offer the greatest advantage.

Discovering new possibilities in quantum computing

The V-score method also reveals which problems are the hardest to solve with existing classical computing methods, which could guide future research. For instance, one-dimensional systems, like chains of particles, are relatively straightforward to solve with current methods, while more complex multidimensional systems, such as quantum lattices, present a far greater challenge. In these areas, quantum computing can offer significant advantages, as new technologies like neural networks and quantum circuits show promising results even compared to existing techniques.

This research not only helps scientists identify the limitations of current methods but also points to the most promising areas for further development in quantum computing. Industries that rely on quantum simulations, such as pharmaceuticals and energy, could leverage these findings to focus on quantum problems that may yield a competitive advantage.

Benefits for future research and industry

The development of quantum computers opens the door to solving problems that are beyond the reach of today's classical computers. Although quantum computing is still in its developmental stage, methods like the V-score allow for more precise identification of problems where this technology could provide key advantages. For example, in the pharmaceutical industry, quantum simulations could accelerate the discovery of new drugs, while in energy, they could aid in the development of new materials with improved properties.

One of the most significant outcomes of this research is the ability to predict where quantum computing, once it matures, will have the greatest impact. By identifying the most complex problems, scientists can direct their efforts toward developing new algorithms and methods that will enable the resolution of these issues. This approach also helps align research with the needs of industry, creating a bridge between the academic community and the business sector.

Source: École Polytechnique Fédérale de Lausanne