CoChat
Sign Up
PreprintCited by 94Quantum Computing Algorithms and ArchitectureQuantum many-body systemsQuantum Information and Cryptography

A universal training algorithm for quantum deep learning

Open access
Verdon, Guillaume, Pye, Jason, Broughton, Michael·2018·arXiv
DOI 10.48550/arxiv.1806.09729arXiv 1806.09729
Open this paper in CoChat — free accountOpen PDF

Abstract

We introduce the Backwards Quantum Propagation of Phase errors (Baqprop) principle, a central theme upon which we construct multiple universal optimization heuristics for training both parametrized quantum circuits and classical deep neural networks on a quantum computer. Baqprop encodes error information in relative phases of a quantum wavefunction defined over the space of network parameters; it can be thought of as the unification of the phase kickback principle of quantum computation and of the backpropagation algorithm from classical deep learning. We propose two core heuristics which leverage Baqprop for quantum-enhanced optimization of network parameters: Quantum Dynamical Descent (QDD) and Momentum Measurement Gradient Descent (MoMGrad). QDD uses simulated quantum coherent dynamics for parameter optimization, allowing for quantum tunneling through the hypothesis space landscape. MoMGrad leverages Baqprop to estimate gradients and thereby perform gradient descent on the parameter landscape; it can be thought of as the quantum-classical analogue of QDD. In addition to these core optimization strategies, we propose various methods for parallelization, regularization, and meta-learning as augmentations to MoMGrad and QDD. We introduce several quantum-coherent adaptations of canonical classical feedforward neural networks, and study how Baqprop can be used to optimize such networks. We develop multiple applications of parametric circuit learning for quantum data, and show how to perform Baqprop in each case. One such application allows for the training of hybrid quantum-classical neural-circuit networks, via the seamless integration of Baqprop with classical backpropagation. Finally, for a representative subset of these proposed applications, we demonstrate the training of these networks via numerical simulations of implementations of QDD and MoMGrad.

Integrity signals

Registry

DataCite · 10.48550/arxiv.1806.09729

Cites

5 of 5 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Claims the paper makes

We couldn’t retrieve this paper’s full text, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Given the close connection between data compression and decoupling, the autoencoder cost function is based on the overlap between the output state on B and a fixed pure state.

    “Given the close connection between data compression and decoupling [163], the cost function is based on the overlap between the output state on B and a fixed pure state.”

    From Variational quantum algorithms (2021) · cites this paper as [163]

  • Ref [163] introduced a VQA for quantum autoencoding with the goal of compressing quantum data.

    “As a quantum analog, Ref. [163] introduced a VQA for quantum autoencoding, with the goal of compressing quantum data.”

    From Variational quantum algorithms (2021) · cites this paper as [163]

  • For ML applications, quantum computers may provide an added benefit since they can avoid getting stuck at relative minima in gradient descent, by quantum tunneling through hills.

    “Moreover, for ML applications, quantum computers may provide an added benefit since they can avoid getting stuck at relative minima in gradient descent, by quantum tunneling through”

    From On Circuit-Based Hybrid Quantum Neural Networks for Remote Sensing Imagery Classification (2022) · cites this paper as [53]

Cites

Flagged first, then by citations

1 of the 6 references couldn’t be matched to a known paper.

Cited by

Sorted by citations

Related

Authors
Verdon, G., Pye, J., & Broughton, M.
ORCID
Venue
arXiv
Publisher
arXiv
Published
2018
ISSN
2331-8422 (ISSN-L)
DOI
10.48550/arxiv.1806.09729
arXiv
1806.09729
OpenAlex
W2949320253
Cited by
68
References
6
Citations by year
2026: 1 · 2025: 3 · 2024: 5 · 2023: 11 · 2022: 9 · 2021: 13 · 2020: 17 · 2019: 8 · 2018: 1
Type
posted content
Language
en
Open access
Licence
arxiv.org/licenses/nonexclusive-distrib/1.0/
Locations
  • arXiv (Cornell University) (submittedVersion)
  • arXiv (Cornell University)
Topics
  • Physical Sciences › Computer Science › Artificial Intelligence › Quantum Computing Algorithms and Architecture
  • Physical Sciences › Physics and Astronomy › Atomic and Molecular Physics, and Optics › Quantum many-body systems
  • Physical Sciences › Computer Science › Artificial Intelligence › Quantum Information and Cryptography
Keywords
quantum deep learning, variational quantum circuits, quantum-enhanced optimization, gradient estimation, quantum tunneling, hybrid quantum-classical neural networks, quantum neural networks, meta-learning, regularization, parallelization

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review