{"id":"arxiv:2610.39847","kind":"preprint","title":"Digital-Analog Quantum Neural Network with Long-Range Interactions","authors":["Zongyuan Ge","Daili Li","Zewen Zhang"],"author_count":3,"year":2026,"published":"2026-10-29","venue":null,"cited_by":null,"doi":null,"arxiv_id":"2610.39847","nct_id":null,"source_url":"https://arxiv.org/abs/2610.39847","archive_url":"https://space-kitty.com/papers/arxiv:2610.39847","updated":"2026-10-29","pmid":null,"pmcid":null,"openalex_id":null,"pdf_url":"https://arxiv.org/pdf/2610.39847","abstract":"Protocols combining single-qubit digital operations with analog quantum evolutions bypass large numbers of two-qubit entangling gates, offering a promising approach to variational algorithms and hybrid quantum simulation. Focusing on the digital-analog quantum neural network (DA-QNN) as a representative framework, this work demonstrates that incorporating long-range interactions enhances the efficiency of such hybrid algorithms. Within the analog blocks, an Ising model with power-law interactions is employed to generate entanglement, and this protocol is benchmarked across supervised and unsupervised learning tasks. Beyond discussing the effects of the transverse field and evolution time, the primary analysis focuses on how the interaction range governs the expressibility and trainability of the ansatz. For supervised image classification and many-body phase detection, long-range ansatzes match the performance of their short-range counterparts with fewer DA-QNN layers or shorter evolution times, owing to the rapid information propagation across the system enabled by nonlocal couplings. This resource reduction alleviates the classical optimization overhead and eases experimental implementation. In unsupervised phase detection, although the long-range ansatz does not outperform its short-range counterpart in every regime, the two ansatzes provide complementary signatures that jointly reconstruct the complete phase diagram. These findings establish long-range interacting platforms as a compelling resource for advancing practical, large-scale digital-analog quantum computing.","abstract_license":"CC0-1.0 (arXiv descriptive metadata)","license":null,"is_open_access":true,"retracted":false,"peer_reviewed":false,"categories":["quant-ph"],"details":null,"provenance":"arxiv","attribution":"Abstract: arXiv descriptive metadata (CC0). Thank you to arXiv for use of its open access interoperability. This service was not reviewed or approved by, nor does it necessarily express or reflect the policies or opinions of, arXiv.","topics":[{"slug":"many-body-quantum-simulation","name":"Quantum Many-Body Physics & Simulation","domain_slug":"physics"},{"slug":"quantum-foundations","name":"Quantum Foundations","domain_slug":"physics"}],"cited_in_dossiers":[]}