Expressive Power and Limitations of Multi-photon Quantum Neural Networks

• URL: http://arxiv.org/abs/2608.01365v1
• Date: Sun, 02 Aug 2026 16:31:28 GMT
• Title: Expressive Power and Limitations of Multi-photon Quantum Neural Networks
• Abstract summary: Multi-photon quantum neural networks (MPQNNs) have the potential to enhance the expressivity through increasing the photon number.<n>Our work elucidates the performance enhancement of multi-photon quantum feature in QNNs, as well as its limitations.
• Score: 19.319091035741742
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Quantum neural networks (QNNs) have shown promise in leveraging quantum computation for machine learning tasks. Utilizing multiple identical photons as input, multi-photon quantum neural networks (MPQNNs) have the potential to enhance the expressivity through increasing the photon number. However, how precisely the expressivity of an MPQNN is affected by an increase in photon number, and whether it can be infinitely enhanced by increasing the photon number, remains unexplored. In this work, we quantitatively estimate the expressivity of this model by deriving upper bounds on approximation error in two cases. In the case of a fixed observable, there exists a threshold that scales linearly with the mode number. Below the threshold, the expressivity of an MPQNN can be enhanced polynomially by increasing the photon number. Above the threshold, however, increasing the photon number does not affect the expressivity. In the case of a trainable observable, the expressivity can always be enhanced polynomially by increasing the photon number. These findings are then validated by numerical simulations. Our work elucidates the performance enhancement of multi-photon quantum feature in QNNs, as well as its limitations, offering guidance for leveraging multi-photon advantages in quantum machine learning.

Related papers

• The trainability of photonic quantum circuits [34.13373752364321]
  Variational quantum algorithms are a leading approach to near-term quantum computing, but their scalability can be limited by barren plateaus and the sampling cost of resolving small changes in the loss landscape.<n>Here, we study the trainability of passive linear-optical quantum circuits and introduce a framework based on the ratio of variance to circuit variance.<n>These results establish photonic variational quantum computing as a promising platform for near-term applications.
  arXiv  Detail & Related papers  (2026-07-23T17:31:09Z)
• Photon Number Coherence of a Quantum Dot-Cavity System Excited Using the SUPER Scheme [0.0]
  Quantum dots embedded in photonic microcavities offer a mature source of single photons.<n>We show that the Swing-UP of quantum EmitteR population (SUPER) scheme can significantly decrease the PNC of the emitted photon, compared to resonant excitation.
  arXiv  Detail & Related papers  (2026-05-04T11:42:06Z)
• Harnessing Photon Indistinguishability in Quantum Extreme Learning Machines [0.723142344259161]
  Quantum extreme machine learning (QELM) protocol leveraging indistinguishable photon pairs and multimode fiber as a random densly connected layer.<n>We experimentally study QELM performance based on photon coincidences -- for distinguishable and indistinguishable photons -- on an image classification task.<n>We relate this improved performance to the enhanced dimensionality and expressivity of the feature space.
  arXiv  Detail & Related papers  (2025-05-16T13:28:01Z)
• Photon-by-photon quantum light state engineering [41.94295877935867]
  We present a review of the progress made over the last few decades in the engineering of quantum light states.<n>Although far from exhaustive, this review aims at providing a sufficiently wide and updated introduction that may serve as the entry point to such a fascinating and rapidly evolving field.
  arXiv  Detail & Related papers  (2025-02-27T16:38:10Z)
• Simulating lossy Gaussian boson sampling with matrix product operators [7.33258560389563]
  We show that efficient tensor network simulations are likely possible under the $N_textoutproptosqrtN$ scaling of the number of surviving photons. We overcome previous challenges due to the large local space dimensions in Gaussian boson sampling with hardware acceleration.
  arXiv  Detail & Related papers  (2023-01-30T12:10:39Z)
• Towards Neural Variational Monte Carlo That Scales Linearly with System Size [67.09349921751341]
  Quantum many-body problems are central to demystifying some exotic quantum phenomena, e.g., high-temperature superconductors. The combination of neural networks (NN) for representing quantum states, and the Variational Monte Carlo (VMC) algorithm, has been shown to be a promising method for solving such problems. We propose a NN architecture called Vector-Quantized Neural Quantum States (VQ-NQS) that utilizes vector-quantization techniques to leverage redundancies in the local-energy calculations of the VMC algorithm.
  arXiv  Detail & Related papers  (2022-12-21T19:00:04Z)
• Quantum density matrix theory for a laser without adiabatic elimination of the population inversion: transition to lasing in the class-B limit [62.997667081978825]
  No class-B quantum density-matrix model is available to date, capable of accurately describing coherence and photon correlations within a unified theory. Here we carry out a density-matrix theoretical approach for generic class-B lasers, and provide closed equations for the photonic and atomic reduced density matrix in the Fock basis of photons. This model enables the study of few-photon bifurcations and non-classical photon correlations in class-B laser devices, also leveraging quantum descriptions of coherently coupled nanolaser arrays.
  arXiv  Detail & Related papers  (2022-05-26T16:33:51Z)
• Parametrized constant-depth quantum neuron [56.51261027148046]
  We propose a framework that builds quantum neurons based on kernel machines. We present here a neuron that applies a tensor-product feature mapping to an exponentially larger space. It turns out that parametrization allows the proposed neuron to optimally fit underlying patterns that the existing neuron cannot fit.
  arXiv  Detail & Related papers  (2022-02-25T04:57:41Z)
• Sequential generation of multiphoton entanglement with a Rydberg superatom [13.103939548290306]
  We experimentally demonstrate an efficient approach for multi-photon generation with a Rydberg superatom. We detect the multiphoton entanglement via converting the photon number degree to a time-bin degree. The efficiency scaling factor for adding one photon is 0.27, surpassing previous results.
  arXiv  Detail & Related papers  (2021-12-17T11:33:11Z)
• Quantum dots as potential sources of strongly entangled photons for quantum networks [0.0]
  A network of quantum repeaters containing multiple sources of entangled photons would allow to overcome a natural limit for transmission distance. Semiconductor quantum dots excel in this context as sub-poissonian sources of polarization entangled photon pairs. We present the state-of-the-art set by GaAs based quantum dots and use them as a benchmark to discuss the challenges to overcome towards the realization of practical quantum networks.
  arXiv  Detail & Related papers  (2020-11-25T13:39:46Z)
• Experimental Quantum Generative Adversarial Networks for Image Generation [93.06926114985761]
  We experimentally achieve the learning and generation of real-world hand-written digit images on a superconducting quantum processor. Our work provides guidance for developing advanced quantum generative models on near-term quantum devices.
  arXiv  Detail & Related papers  (2020-10-13T06:57:17Z)
• A bright and fast source of coherent single photons [46.25143811066789]
  A single photon source is a key enabling technology in device-independent quantum communication. We report a single photon source with an especially high system efficiency.
  arXiv  Detail & Related papers  (2020-07-24T17:08:46Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.