Quantum computers are supposed to revolutionize scientific computing, but researchers have recently shown that a classical computer can sometimes outperform them. Using advanced mathematical models, a team solved a complex quantum physics problem, initially thought to be inaccessible without a classical computer.
The key to this success lies in a phenomenon called "confinement," which restricts the spread of entanglement effects within the system. This discovery, published in
Physical Review Letters, sheds new light on the blurry boundary between the performance of classical and quantum computers.
Joseph Tindall, a researcher at the Flatiron Institute, and his team studied a model of magnetic spins in a two-dimensional lattice. In this model, each "spin" acts as a tiny magnet that can point either up or down, or even exist in a superposition of both states. When a magnetic field is applied, the spins start to interact with each other, producing entanglement, that is, a quantum link between their states.
However, Tindall observed that this entanglement remained limited to small groups of neighboring spins, rather than spreading throughout the system. This confinement reduces the complexity of the calculation, making it solvable by a classical computer. This unexpected simplification is the result of the system's specific architecture, which channels the entanglement effects in a localized manner.
IBM had designed a complex magnetic simulation problem to test its quantum computers. According to their researchers, this calculation should have been impossible for a classical computer. But in two weeks, Tindall proved he could crack it using a classical model, and even with the capabilities of a smartphone.
His approach relies on classical techniques that, while known, are rarely applied in quantum simulation. Through an ingenious assembly of methods, Tindall demonstrated that the entanglement confinement simplified the problem to the point where it could be solved without quantum technologies.
Confinement, in this context, works similarly to quark confinement in particle physics. It means that the spins in the system remain mostly aligned in an orderly fashion and do not become chaotic. This behavior, far from what is expected from a "free" quantum system, limits the entanglement and thus reduces the computational complexity.
This discovery opens up new perspectives for assessing where quantum computers might truly outperform classical ones.
The algorithms developed by Tindall and his colleagues could, therefore, become reference tools for future experiments. This study represents a step forward in clearly delineating the boundary between the capabilities of classical and quantum computers.