Q-Capsule: A Localized Capsule-Based Quantum Neural Architecture for Barren Plateau Mitigation
Abstract
Variational quantum algorithms are often limited by barren plateaus: gradients vanish as circuit size and depth increase, making quantum neural networks difficult to train. We propose Q-Capsule, a localized capsule-based quantum neural architecture that mitigates this problem through register partitioning, local readout, sparse inter-capsule coupling, trainable data re-uploading, and Quantum Fisher Information Matrix (QFIM)-guided adaptive depth growth. By restricting the dominant support of each observable to a small capsule and controlling inter-capsule entanglement, Q-Capsule preserves useful gradient signals while retaining communication between local quantum representations. As the register width increases, Q-Capsule consistently maintains stable gradient variance, whereas globally entangling baselines exhibit exponential suppression with a log-gradient-variance slope near -ln 2 per qubit. Q-Capsule also produces more structured optimization landscapes, higher parameter efficiency, improved robustness to depolarizing noise, and lower measurement requirements. Its adaptive policy achieves 98.1% accuracy on binary classification and 97.7% on four-class classification, while using approximately 73% fewer two-qubit gates than the fixed-deep model on the multiclass task.
Related
- Scalable Megapixel resolution Solid-State LiDAR architecture by Dual-MEMS coarse-fine Diffractive Beam Steering with Crosstalk Mitigation
- Quantum error mitigation simulates general non-Hermitian dynamics
- Plateau-Constrained Selection: Exploiting Degeneracy for Lower-Depth Quantum Compilation
- Feynman's clock and hierarchy-informed sampling for quantum error mitigation
- Universality of the 1/9 Magnetization Plateau and Quantum-Disordered States in the Kagome Family Cs8AB3Ti12F48 (A=Rb,Li; B=K,Na)
BibTeX
@misc{fime2026qcapsule,
title = {{Q-Capsule: A Localized Capsule-Based Quantum Neural Architecture for Barren Plateau Mitigation}},
author = {Awal Ahmed Fime and Tasfia Zaman Samiha and Saika Zaman and Dimitris Pados and George Sklivanitis and Abdur R. Shahid and Ahmed Imteaj},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.11261},
eprint = {2610.11261},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.11261},
}
RIS
TY - UNPB
TI - Q-Capsule: A Localized Capsule-Based Quantum Neural Architecture for Barren Plateau Mitigation
AU - Awal Ahmed Fime
AU - Tasfia Zaman Samiha
AU - Saika Zaman
AU - Dimitris Pados
AU - George Sklivanitis
AU - Abdur R. Shahid
AU - Ahmed Imteaj
PY - 2026
DA - 2026/10/08
PB - arXiv
DO - 10.48550/arxiv.2610.11261
UR - https://arxiv.org/abs/2610.11261
AB - Variational quantum algorithms are often limited by barren plateaus: gradients vanish as circuit size and depth increase, making quantum neural networks difficult to train. We propose Q-Capsule, a localized capsule-based quantum neural architecture that mitigates this problem through register partitioning, local readout, sparse inter-capsule coupling, trainable data re-uploading, and Quantum Fisher Information Matrix (QFIM)-guided adaptive depth growth. By restricting the dominant support of each observable to a small capsule and controlling inter-capsule entanglement, Q-Capsule preserves useful gradient signals while retaining communication between local quantum representations. As the register width increases, Q-Capsule consistently maintains stable gradient variance, whereas globally entangling baselines exhibit exponential suppression with a log-gradient-variance slope near -ln 2 per qubit. Q-Capsule also produces more structured optimization landscapes, higher parameter efficiency, improved robustness to depolarizing noise, and lower measurement requirements. Its adaptive policy achieves 98.1% accuracy on binary classification and 97.7% on four-class classification, while using approximately 73% fewer two-qubit gates than the fixed-deep model on the multiclass task.
ER -