Dimension-adaptive machine learning-based quantum state reconstruction

We introduce an approach for performing quantum state reconstruction on systems of n qubits using a machine learning-based system trained exclusively m qubits, where ≥ n. This removes the necessity exactly matching dimensionality under consideration with dimension model used training. demonstrate our technique by randomly sampled one, two, and three methods containing at least one additional qubit. The time required scales significantly more favorably than training time; hence this can offer overall saving resources leveraging single neural network dimension-variable reconstruction, obviating need to train dedicated learning each Hilbert space.