ECAMP 15

Quantum control of a wide class of molecules is crucial for advancing a variety of quantum applications. The potential of polyatomic molecular ions in this context can be significantly enhanced using the toolbox provided by quantum information processing and quantum logic spectroscopy (QLS).
These techniques rely on the ability to determine the state of the molecule which is almost impossible to track in some molecular species, such as the CaOH⁺ molecule, where the presence of numerous states, closely spaced energy levels, and overlapping transitions make direct state readout difficult.
Quantum logic spectroscopy enables precise interrogation of molecular states by leveraging co-trapped atomic ions. We explore Bayesian state estimation to infer the molecular state distribution based on measurement outcomes, incorporating prior knowledge from the state distribution of the molecule.
To enhance the efficiency of state identification, we implement Bayesian adaptive design, where the next measurement is selected to maximize the expected information gain. In addition, we will compare our simulations with experimental data, refining our understanding of dynamics of the molecular rotation under QLS protocols. Bayesian adaptive design comes with a considerable computational cost but can be made more efficient by the application of deep experiment design, training a neural network to simulate experimental outcomes and optimize measurement strategies. This approach will allow us to make real-time decisions during spectroscopy experiments, dynamically adjusting measurement choices for optimal information extraction.