Anastasia Pipi | Searcharxiv

arXiv2026

Inverse Design of Quantum Control Sequences with Fourier Neural Operators

Quantum optimal control is a key tool for steering quantum dynamics, but its computational cost grows rapidly with the Hilbert space dimension. Here, we introduce a Fourier Neural Operator (FNO)-based framework for learning high dimensional molecular quantum dynamics and accelerating the inverse design of control protocols. Given an initial molecular population distribution, laser frequency, and polarization, the FNO predicts molecular-motional population dynamics up to $10^7$ times faster than GPU-accelerated numerical propagation with CUDA-Q Dynamics. Using this fast and differentiable surrogate, we develop the FNO stochastic pulse-measurement planner (FNO-SPMP), which constructs pulse sequences to purify an initially mixed Boltzmann distribution. We demonstrate the protocol in an 888-dimensional subspace of the hydronium molecule at 20 K, achieving a target-state population of 0.98 with a sequence success rate of up to 86.2%. In a shared discrete control space, FNO-SPMP achieves nearly twice the success rate of a reinforcement-learning baseline while using roughly half as many quantum control pulses and reducing pulse-sequence generation time from approximately 10 hours to 10-20 minutes. These results show that operator-learning surrogates can enable inverse design in quantum systems whose Hilbert spaces are too large for conventional direct optimization.

quant-ph↗

arXiv2026

Molecular Quantum Control Algorithm Design by Reinforcement Learning

Precision measurements of molecules offer an unparalleled paradigm to probe physics beyond the Standard Model. The rich internal structure within these molecules makes them exquisite sensors for detecting fundamental symmetry violations, local position invariance, and dark matter. While trapping and control of diatomic and a few very simple polyatomic molecules have been experimentally demonstrated, leveraging the complex rovibrational structure of more general polyatomics demands the development of robust and efficient quantum control schemes. In this study, we present reinforcement-learning quantum-logic spectroscopy (RL-QLS), a general, reinforcement-learning-designed, quantum logic approach to prepare molecular ions in single, pure quantum states. The reinforcement learning agent optimizes the pulse sequence, each followed by a projective measurement, and probabilistically manipulates the collapse of the quantum system to a single state. The performance of the RL-QLS control algorithm is numerically demonstrated for the polyatomic molecule H$_3$O$^+$ with 130 thermally populated eigenstates and degenerate transitions within inversion doublets, where quantum Markov decision process modeling and a physics-informed reward function play a key role, as well as for CaH$^+$ under the disturbance of environmental thermal radiation. The developed theoretical framework cohesively integrates techniques from quantum chemistry, AMO physics, and artificial intelligence, and we expect that the results can be readily implemented for quantum control of polyatomic molecular ions with densely populated structures, thereby enabling experimental tests of fundamental theories.

quant-ph↗