Curator's Take
AI Commentary
This article spotlights a hidden but critical failure mode that emerges when quantum neural‑network code written for simulators is run on real devices with hundreds of physical qubits: the SamplerQNN post‑processing filter discards most measurement shots, leaving an unnormalised probability vector and dramatically degrading both training gradients and inference accuracy. By quantifying losses up to 99 % and showing a 22–27× compression of loss signals across multiple IBM backends, the work underscores how software assumptions about virtual qubit layouts can cripple scalability unless the classical stack is co‑designed with hardware realities. The authors’ layout‑aware marginalisation fix, now merged into Qiskit‑ML, provides an immediate remedy and a cautionary template for other quantum‑ML libraries as the field moves toward utility‑scale processors.
— Mark Eatherly
Summary
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.