Loading
Life sciences · Preprint
arXiv · September 4, 2026
Early or partial results. Treat as a signal, not a conclusion.
This preprint documents a critical data loss bug in Qiskit Machine Learning's SamplerQNN class, where postprocessing filters discard 85–99.6% of valid measurement shots on utility-scale quantum hardware, causing inference accuracy to drop from 0.94 to 0.39 and compressing the training loss signal by 22–27×. The authors present a marginalisation-based fix merged into the codebase but do not report independent peer review or validation beyond the two IBM backends tested.
Case study with empirical quantification on two quantum hardware backends. Quantum neural network experiments using Qiskit Machine Learning SamplerQNN class on IBM quantum hardware; no patient or human subject population.. Intervention: Layout-based marginalisation fix (Pull Request #1041) to postprocessing routine. Compared with: Original postprocessing method in library versions 0.8.4 to 0.9.0. IBM quantum backends (specific centres not named).
Postprocessing filter discards 85 to 99.6% of valid measurement shots depending on qubit placement strategy Inference accuracy drops from 0.94 to 0.39 on the same raw measurements due to data loss Training loss signal compressed by 22 to 27× across experiments, reducing optimizer sensitivity
Safety was not reported in the material analysed. Check the source before drawing any conclusion about harm.
The source did not state who this applies to in practice.
A case study identifying a data loss bug in production quantum machine learning software, with quantified impact on inference and training but no controlled experiments, peer review, or validation beyond two IBM backends.
As stated by the source record.
Quoted from the source exactly as published.
Graded across the dimensions that decide whether you should act, each from what the source actually supports. There is no single score, and where a dimension was not assessed it says so.
As quantum hardware scales to larger devices, the classical software layers that interface with it must evolve in step. Postprocessing routines developed and tested primarily in simulator settings can encode assumptions that no longer hold on utility-scale devices, leading to data loss that can be difficult to detect from high-level model outputs alone. We present a case study of \texttt{SamplerQNN}, the sampling-based quantum neural network class in the Qiskit Machine Learning library. Here, the postprocessing method applies a filter that assumes measurement bit-strings are in virtual qubit space. On our quantum hardware runs, where bit-strings span over 100 physical qubits, this filter led to the loss of 85 to 99.6\% of valid measurement shots, depending on the transpiler's qubit placement. The resulting probability vector is unnormalised, allowing distorted prediction and loss values to propagate through the model without an API-level warning. We demonstrate the impact across five experiments on two IBM backends: for inference, accuracy drops from 0.94 to 0.39 on the same raw measurements; for training, the loss signal is compressed by 22 to 27$\times$, substantially reducing the sensitivity of the optimiser to the objective landscape. The behaviour arises in all released versions of the library (0.8.4 to 0.9.0). We implemented a layout-based marginalisation fix, merged into the GitHub codebase as Pull Request \#1041, that makes \texttt{SamplerQNN} postprocessing forward-compatible with current and upcoming hardware.
Taken from the source record, never inferred. Follow any of these and new work involving them reaches your briefing.