Skip to content
DraconiticResearch
Search Sources API
203,234 studies31 sourcesUpdated 47 min ago
PreprintCC BY 4.0arXiv

Quantum data loading from the learned shared structure of real signals

  • Pablo Herrero Gómez
  • Antonio Jimeno Morenilla
  • David Muñoz-Hernández
  • Higinio Mora Mora
Open at sourcePDFDOICite

Abstract

Preparing quantum states from classical data can cost more than the computation they serve; most loaders tailor a circuit to each input. Here we show that the signals of a real dataset share structure that can be learned once and reused. Our quantum-native loader learns a low-dimensional description of a dataset and prepares every signal with one fixed circuit set by a few numbers. Across seven views of five public datasets it meets the targets of the strongest structured loader at equal gate cost with several times fewer numbers per signal. These numbers can be inferred from a random subset: in a preregistered blind replication the subset needed to come within ten per cent of full-signal accuracy stayed constant within a prespecified margin as signals grew sixteenfold, whereas the structured loader needed ever more. It declines what it cannot represent, covering fewer cases than that baseline and no electrocardiogram.

Related

  1. The Structure of Shared Classical Facts in Quantum Systems
    Jie Gu
    arXivPreprintCC BY 4.0Full textQuantum physics & quantum informationarXiv
  2. Quantum Metrology for Signals with an Unknown Causal Structure
    Gwon Ryul Han, Hyukgun Kwon, Seok Hyung Lie
    arXivPreprintCC BY-NC-SA 4.0Full textQuantum physics & quantum informationarXiv
  3. Readable Is Not Steerable: Pre-Registered Tests of Physical Structure in Learned Fluid Representations
    Manvendra Modgil
    arXivPreprintCC BY 4.0Full textFluid dynamicsarXiv
  4. Mitigating Capacitive Loading Enables Fast Two-Qubit Gates in Highly Connected Fluxonium Quantum Processors
    Quan Guan, Guo Xuan Chan, Xu Dou +2
    arXivPreprintarXiv licenceFull textQuantum physics & quantum informationarXiv
  5. Learned Diffractive Optics for Quantum-Optimal Inference
    Matthew J. Filipovich, Alexander Duplinskii, A. I. Lvovsky
    arXivPreprintCC BY 4.0Full textQuantum physics & quantum informationarXiv

Cite

BibTeXRIS
BibTeX
@misc{gomez2026quantum,
  title = {{Quantum data loading from the learned shared structure of real signals}},
  author = {Pablo Herrero Gómez and Antonio Jimeno Morenilla and David Muñoz-Hernández and Higinio Mora Mora},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2610.06076},
  eprint = {2610.06076},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.06076},
  note = {CC BY 4.0},
}
RIS
TY  - UNPB
TI  - Quantum data loading from the learned shared structure of real signals
AU  - Pablo Herrero Gómez
AU  - Antonio Jimeno Morenilla
AU  - David Muñoz-Hernández
AU  - Higinio Mora Mora
PY  - 2026
DA  - 2026/10/06
PB  - arXiv
DO  - 10.48550/arxiv.2610.06076
UR  - https://arxiv.org/abs/2610.06076
AB  - Preparing quantum states from classical data can cost more than the computation they serve; most loaders tailor a circuit to each input. Here we show that the signals of a real dataset share structure that can be learned once and reused. Our quantum-native loader learns a low-dimensional description of a dataset and prepares every signal with one fixed circuit set by a few numbers. Across seven views of five public datasets it meets the targets of the strongest structured loader at equal gate cost with several times fewer numbers per signal. These numbers can be inferred from a random subset: in a preregistered blind replication the subset needed to come within ten per cent of full-signal accuracy stayed constant within a prespecified margin as signals grew sixteenfold, whereas the structured loader needed ever more. It declines what it cannot represent, covering fewer cases than that baseline and no electrocardiogram.
ER  -