Gast ::
| Hauptseite > Publikationsdatenbank > Efficient Quantum Convolutional Neural Networks for Image Classification: Overcoming Hardware Constraints |
| Hauptseite > Publikationsdatenbank > Efficient Quantum Convolutional Neural Networks for Image Classification: Overcoming Hardware Constraints |
| Poster (After Call) | FZJ-2025-05552 |
; ; ; ;
2025
This record in other databases:
Please use a persistent id in citations: doi:10.34734/FZJ-2025-05552
Abstract: While classical convolutional neural networks (CNNs) have revolutionized imageclassification, the emergence of quantum CNNs (QCNNs) presents new opportunities with high-dimensional transformations that are infeasible for classicalCNNs. However, their implementation on current noisy intermediate-scale quantum (NISQ) devices remains challenging due to hardware limitations. In ourresearch, we address this challenge by introducing fragment encoding that significantly reduces the input dimensionality. We show that accuracy improvesas the number of qubits increases from 1 to 4 to 16, providing evidence offavorable scaling. Motivated by this, we developed a primitive QCNN architecture with 49 qubits which is sufficient to directly process 28 × 28 pixelMNIST images, eliminating the need for classical dimensionality reductionpre-processing. Additionally, we propose an automated framework based onexpressibility, entanglement, and complexity characteristics to identify the building blocks of QCNNs, parameterized quantum circuits (PQCs). Our approachdemonstrates advantages in convergence speed with a similar parameter countcompared to both hybrid QCNNs and classical CNNs. We validated our experiments on IBM’s Heron r2 quantum processor, achieving 96.08% accuracy withapproximately 70% faster convergence relative to classical CNNs under identicaltraining conditions. Our results show that simple image-classification tasks canbe realized on real quantum hardware today.
|
The record appears in these collections: |
PDF