The quantum Mpemba effect is characterized by exponential relaxation from a nonequilibrium state to a stationary one. It was previously predicted by analysis of the Liouville superoperator and experimentally demonstrated in a three-level system. In this work, the system's dynamics within this effect are described using quantum thermodynamics of maximum entropy growth for an isolated three-level system. To compare theory with experiment, the system's Hilbert space is projected from four dimensions to three using Feshbach projection. The relaxation parameter τ_D, which determines the dissipative acceleration, is found using machine learning methods. The result is a model that thermodynamically describes this phenomenon at the quantum level.
In the quantum world, the path to calm sometimes resembles a paradoxical race: starting further from the finish, the system reaches equilibrium faster. This effect, a quantum analog of the Mpemba phenomenon, is due to the rapid growth of entropy — a measure of disorder. A strong push at the start seems to trigger an avalanche of changes that accelerates settling.
Scientists built a model based on the principle of maximum entropy growth, tracing back to Ludwig Boltzmann. By compressing the description from four to three dimensions, they achieved agreement with a laboratory three-level system. Machine learning revealed the key parameter governing the decay rate. Thus, a precise formula for the quantum 'shake' emerged.
This is not just theory: qubits will be able to reach the desired state faster after a short excitation. This engineering trick will shorten waiting times in future devices.
🎯 Aristotle described the rapid freezing of heated water, but the effect was named after Tanzanian schoolboy Erasto Mpemba: in 1963, he noticed that hot ice cream freezes faster.