Automated Optimization of Laser Fields for Quantum State Manipulation

Abstract

A gradient-based optimization approach combined with automatic differentiation is employed to ensure high accuracy and scalability when working with high-dimensional parameter spaces. Numerical simulations confirm the effectiveness of the proposed method: the population is reliably transferred to the target state with minimal occupation of intermediate levels, while the control pulses remain smooth and physically implementable. The developed framework serves as a universal and experimentally applicable tool for automated control pulse design in quantum systems. It is particularly useful in scenarios where analytical methods or manual parameter tuning--such as standard schemes like STIRAP--prove to be inefficient or inapplicable.