Optimising Quantum Calculations Reliability via Machine Learning: The IBM Case Study.

Abstract

The current quantum technology depends on hardware/time-dependent features that render quantum calculations highly error-prone. Having faulty calculations generally requires redoing them, which induces an economic/computational overhead by using quantum machines/simulators. This can be problematic due to unaffordable fees, machine unavailability, or loss of quantum advantage. An alternative can be performing the calculations when the machines are in their most suitable state. This work presents a pipeline based on Machine Learning (ML), allowing users to choose the appropriate moment to perform a given computation based on the estimation of the Jensen-Shannon divergence between the noisy and ideal distributions of quantum sampling. This includes (I) an extract-transform-load data module, (II) an ML unit for quantum features forecasting and error prediction, and (III) a web-based visualisation unit. The pipeline was built/tested using 3.5 months of calibration data from three real 127-qubit IBM quantum machines. The results confirmed the applicability of the proposal in realistic scenarios.