Solving equations that govern dissipative quantum systems is a complex task. However, an artificial intelligence-based surrogate model has been introduced to simplify this process. The model uses Fourier neural operators to parameterize quantum propagators, which are then trained using dataset and physics-informed loss functions. This approach eliminates the need for time-consuming iterations and provides a universal super-operator that can evolve any initial quantum state for arbitrarily long times. The model has been tested on the Fenna-Matthews-Olson complex, a pigment-protein complex found in green sulfur bacteria, demonstrating its potential for improving simulation methods.
↧