Dimension-adaptive machine-learning-based quantum state reconstruction
This addresses a practical bottleneck in quantum information processing by reducing training overhead for dimension-variable state reconstruction, though it appears incremental as an extension of existing machine-learning methods.
The paper tackles the problem of quantum state reconstruction for n-qubit systems by introducing a machine-learning approach trained on m-qubit systems where m≥n, eliminating the need for dimension-matched training. They demonstrated this on 1-3 qubit systems using models trained with at least one extra qubit, achieving resource savings by using a single neural network for variable dimensions.
We introduce an approach for performing quantum state reconstruction on systems of $n$ qubits using a machine-learning-based reconstruction system trained exclusively on $m$ qubits, where $m\geq n$. This approach removes the necessity of exactly matching the dimensionality of a system under consideration with the dimension of a model used for training. We demonstrate our technique by performing quantum state reconstruction on randomly sampled systems of one, two, and three qubits using machine-learning-based methods trained exclusively on systems containing at least one additional qubit. The reconstruction time required for machine-learning-based methods scales significantly more favorably than the training time; hence this technique can offer an overall savings of resources by leveraging a single neural network for dimension-variable state reconstruction, obviating the need to train dedicated machine-learning systems for each Hilbert space.