Soft Mixture of Quantum Experts for Hybrid Quantum-Classical Learning
Organizations: National Research Council of Italy (CNR), Institute of Informatics and Telematics (IIT), via Giuseppe Moruzzi 1, Pisa, Italy · Computational Biomedicine Unit, Department of Medical Sciences, University of Turin, via Santena 19, Torino, Italy
Abstract
Quantum Machine Learning (QML) has attracted increasing interest as a possible route to construct feature representations that are difficult to reproduce with classical models. However, near-term QML architectures based on trainable parametrized quantum circuits remain affected by limited hardware resources, noise, and severe optimization difficulties such as barren plateaus. In this work we propose Soft-MoQE, a Soft Mixture of Quantum Experts architecture that replaces variational quantum training with the input-dependent selection and combination of fixed quantum feature maps. The model is composed of a heterogeneous pool of non-trainable quantum circuits, each defining a distinct embedding, and a classical gating network that learns how to weight these embeddings for each data point. We benchmark Soft-MoQE on three PCA-reduced tasks: MNIST parity, MNIST 10-class, and FashionMNIST. Soft-MoQE outperforms the single best expert, the uniform average of all experts and a linear baseline, achieving task-dependent performance comparable to that of a trainable quantum neural network, without optimizing any quantum parameter and with a quantum training cost that is lower by orders of magnitude. The analysis of the class- conditioned expert usage shows that the learned routing combines a few experts for each input, with a selection that depends on the class. These results support the learned routing among fixed quantum circuits as a trainability-aware design principle for hybrid QML.
Figures & tables
| Family | Entanglement | Axis | Experts | |
|---|---|---|---|---|
| Rotational | none | 2, 3, 4 | 3 | |
| Hadamard | none | 2, 3, 4 | 3 | |
| 3 | 1 | |||
| IQP | linear | 2, 3, 4 | 3 | |
| full | 2, 3, 4 | 3 | ||
| Entangled rotational | CNOT ladder | , | 2, 3, 4 | 6 |
| Model | MNIST: Parity | MNIST: 10-class | FashionMNIST | |||
|---|---|---|---|---|---|---|
| MCC | Accuracy | MCC | Accuracy | MCC | Accuracy | |
| Single best expert | ||||||
| Uniform average of experts | ||||||
| Soft-MoQE | ||||||
| Model | MNIST: Parity | MNIST: 10-class | FashionMNIST | |||
|---|---|---|---|---|---|---|
| MCC | Accuracy | MCC | Accuracy | MCC | Accuracy | |
| Soft-MoQE | ||||||
| QNN | ||||||
| Linear baseline | ||||||
| MLP | ||||||
Appendix figures & tables2 assets
Supplementary material from the paper’s appendix.
Appendix
Hybrid Quantum-Classical ML
Hybrid quantum-classical machine learning combines quantum circuits or annealers with classical models for learning, prediction, generation, and optimization.
See all 198 papers