Implementation of Active Learning (AL) Quantum State Tomography for various quantum states generated on IBMs quantum devices or with DMRG. Our AL scheme decides, based on previous measurements, which meansurement configuration should be considered in the next measurement. In order to find the measurement configuration that will improve the learned state representation most efficiently, we apply the query-by-committee strategy, where the members of the committee are represented by restricted Boltzmann machine (RBM) quantum states. The complete algorithm is shown below. See our paper for further details: Lange2023, Quantum 7.
QuCumber for the RBM reconstruction.
An example can be run by calling example.py.