CohAlign: Predicting the first-order noise susceptibility of gradients in equivariant quantum neural networks
- 1Centre for Visual Computing and Intelligent Systems, University of Bradford, Bradford, United Kingdom.
- 2School of Individualized Study, Rochester Institute of Technology, Rochester, New York, United States of America.
Abstract
Quantum neural networks learn by following gradients, and on real hardware those gradients are washed out by noise. Predicting how quickly this happens, before any finite noise experiment or circuit sweep is run, remains an open problem. We present CohAlign, an open source toolkit that computes the first order noise susceptibility of the ensemble second moment of a specified parameter gradient in a charge conserving brickwork quantum neural network. The computation needs only a classical description of the circuit, the noise model, the prepared state, the readout and the initialisation prior. The predicted quantity is distinct from training convergence or final accuracy. The key observation is that only the part of the noise that a measurement can actually see matters. A purposefully constructed control channel makes this clear, damaging the operator carrying a gradient at nearly the maximal rate while the measured gradient remains exactly as predicted and untouched. We validate the predictions against separately evaluated finite difference slopes across the benchmark suite, including a signed coherent case, in exact simulations at eight and ten qubits and on two circuit topologies.
Related
- Shaping Slow Manifolds in Equivariant Recurrent Neural Networks
- Petimot: A Novel Framework for Inferring Protein Motions from Sparse Data Using SE(3)-Equivariant Graph Neural Networks
- Phytoplankton Dynamics and Urbanization Gradients Reshape Dissolved Organic Matter and Microbial Networks across a Freshwater Estuary
- Predicting flux-partitioning error in quasi-steady-state reductions of branched biochemical networks
- SPECTRA: predicting cellular perturbation responses with Graph Learning over Gene Regulatory Networks
BibTeX
@article{ugail2026cohalign,
title = {{CohAlign: Predicting the first-order noise susceptibility of gradients in equivariant quantum neural networks}},
author = {Hassan Ugail and Newton Howard},
journal = {PLOS One},
year = {2026},
month = oct,
doi = {10.1371/journal.pone.0361139},
url = {https://doi.org/10.1371/journal.pone.0361139},
note = {CC BY},
}
RIS
TY - JOUR
TI - CohAlign: Predicting the first-order noise susceptibility of gradients in equivariant quantum neural networks
AU - Hassan Ugail
AU - Newton Howard
PY - 2026
DA - 2026/10/22
JO - PLOS One
DO - 10.1371/journal.pone.0361139
UR - https://doi.org/10.1371/journal.pone.0361139
AB - Quantum neural networks learn by following gradients, and on real hardware those gradients are washed out by noise. Predicting how quickly this happens, before any finite noise experiment or circuit sweep is run, remains an open problem. We present CohAlign, an open source toolkit that computes the first order noise susceptibility of the ensemble second moment of a specified parameter gradient in a charge conserving brickwork quantum neural network. The computation needs only a classical description of the circuit, the noise model, the prepared state, the readout and the initialisation prior. The predicted quantity is distinct from training convergence or final accuracy. The key observation is that only the part of the noise that a measurement can actually see matters. A purposefully constructed control channel makes this clear, damaging the operator carrying a gradient at nearly the maximal rate while the measured gradient remains exactly as predicted and untouched. We validate the predictions against separately evaluated finite difference slopes across the benchmark suite, including a signed coherent case, in exact simulations at eight and ten qubits and on two circuit topologies.
SN - 1932-6203
ER -