Universality of Classically Trainable, Quantum-Deployed Boson-Sampling Generative Models

• URL: http://arxiv.org/abs/2603.11014v1
• Date: Wed, 11 Mar 2026 17:41:12 GMT
• Title: Universality of Classically Trainable, Quantum-Deployed Boson-Sampling Generative Models
• Abstract summary: We introduce the Boson Sampling Born Machine (BSBM)<n>We show that BSBMs can be trained classically for wide families of loss functions.<n>We argue that "basic" BSBMs are not universal generative models.
• Score: 1.7753787132550425
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent work on the instantaneous quantum polynomial-time (IQP) quantum-circuit Born machine (QCBM) highlights a promising paradigm for generative modeling: train classically, deploy quantumly. In this setting, the training objective can be evaluated efficiently on a classical computer, while sampling from the resulting model may still be classically intractable. Furthermore, in the IQP-QCBM framework, extending the model family with ancillary qubits has been proven to yield universality. This paper asks whether similar results hold for linear-optical generative models. To this end, we introduce the Boson Sampling Born Machine (BSBM). Our analysis retraces analogous steps as were found for IQP-QCBMs with twists. Using recent results that enable classical approximation of broad classes of expectation values in linear optics, we show that BSBMs can be trained classically for wide families of loss functions. Next, we argue that "basic" BSBMs are not universal generative models, and that universality can be achieved by expanding the model while preserving efficient classical training and sampling hardness. In our approach, we introduce and analyze the role of constant-function postprocessing, generalizing the construction for IQP-QCBMs, which under suitable conditions can lead to universality while preserving the hardness of classically simulating the models. We showcase a family of BSBMs, characterized by a single hyperparameter, that allows for a monotonic increase in expressivity toward universality while retaining the capacity to represent ostensibly hard distributions. Furthermore, we discuss the possible modalities for the efficient classical training, in the sense of efficient estimation of gradients of the loss function.

Related papers

• Learnable yet not simulable: a quantum resource theory of learning models [57.695822250391906]
  It remains unclear which quantum resource governs the classical learnability of quantum circuits.<n>We study the expectation-value functions of families of tunable quantum circuits.<n>These results establish a quantitative resource-theoretic framework for delineating the boundary between classical simulation and learning.
  arXiv  Detail & Related papers  (2026-08-03T14:48:03Z)
• Generative modeling with Gaussian Boson Sampling: classically trainable Bosonic Born Machines [0.1449880056570423]
  We introduce a photonic quantum generative model built on parametrized Gaussian Boson Sampling circuits.<n>The training is based on the efficient classical evaluation of expectation values enabled by the Gaussian structure of the state.<n>We demonstrate the effectiveness of the approach through numerical experiments on photonic systems with up to 805 modes and over a million trainable parameters.
  arXiv  Detail & Related papers  (2026-03-11T18:04:11Z)
• Fermionic Born Machines: Classical training of quantum generative models based on Fermion Sampling [0.0]
  We introduce Fermionic Born Machines as an example of classically trainable quantum generative models.<n>The model employs parameterized magic states and fermionic linear optical (FLO) transformations with learnable parameters.<n>The specific structure of the ansatz induces a loss landscape that exhibits favorable characteristics for optimization.
  arXiv  Detail & Related papers  (2025-11-17T19:03:03Z)
• Quantum Latent Diffusion Models [65.16624577812436]
  We propose a potential version of a quantum diffusion model that leverages the established idea of classical latent diffusion models.<n>This involves using a traditional autoencoder to reduce images, followed by operations with variational circuits in the latent space.<n>The results demonstrate an advantage in using a quantum version, as evidenced by obtaining better metrics for the images generated by the quantum version.
  arXiv  Detail & Related papers  (2025-01-19T21:24:02Z)
• Learning hard distributions with quantum-enhanced Variational Autoencoders [2.545905720487589]
  We introduce a quantum-enhanced VAE (QeVAE) that uses quantum correlations to improve the fidelity over classical VAEs. We empirically show that the QeVAE outperforms classical models on several classes of quantum states. Our work paves the way for new applications of quantum generative learning algorithms.
  arXiv  Detail & Related papers  (2023-05-02T16:50:24Z)
• A Framework for Demonstrating Practical Quantum Advantage: Racing Quantum against Classical Generative Models [62.997667081978825]
  We build over a proposed framework for evaluating the generalization performance of generative models. We establish the first comparative race towards practical quantum advantage (PQA) between classical and quantum generative models. Our results suggest that QCBMs are more efficient in the data-limited regime than the other state-of-the-art classical generative models.
  arXiv  Detail & Related papers  (2023-03-27T22:48:28Z)
• A supplemental investigation of non-linearity in quantum generative models with respect to simulatability and optimization [1.4329826483358277]
  We investigate two questions of relevance to the quantum algorithms and machine learning communities. Does introducing this form of non-linearity make the learning model classically simulatable due to the deferred measurement principle? And does introducing this form of non-linearity make the overall model's training more unstable?
  arXiv  Detail & Related papers  (2023-02-01T22:48:26Z)
• Do Quantum Circuit Born Machines Generalize? [58.720142291102135]
  We present the first work in the literature that presents the QCBM's generalization performance as an integral evaluation metric for quantum generative models. We show that the QCBM is able to effectively learn the reweighted dataset and generate unseen samples with higher quality than those in the training set.
  arXiv  Detail & Related papers  (2022-07-27T17:06:34Z)
• Generalization Metrics for Practical Quantum Advantage in Generative Models [68.8204255655161]
  Generative modeling is a widely accepted natural use case for quantum computers. We construct a simple and unambiguous approach to probe practical quantum advantage for generative modeling by measuring the algorithm's generalization performance. Our simulation results show that our quantum-inspired models have up to a $68 times$ enhancement in generating unseen unique and valid samples.
  arXiv  Detail & Related papers  (2022-01-21T16:35:35Z)
• Error mitigation and quantum-assisted simulation in the error corrected regime [77.34726150561087]
  A standard approach to quantum computing is based on the idea of promoting a classically simulable and fault-tolerant set of operations. We show how the addition of noisy magic resources allows one to boost classical quasiprobability simulations of a quantum circuit.
  arXiv  Detail & Related papers  (2021-03-12T20:58:41Z)
• Enhancing Generative Models via Quantum Correlations [1.6099403809839032]
  Generative modeling using samples drawn from the probability distribution constitutes a powerful approach for unsupervised machine learning. We show theoretically that such quantum correlations provide a powerful resource for generative modeling. We numerically test this separation on standard machine learning data sets and show that it holds for practical problems.
  arXiv  Detail & Related papers  (2021-01-20T22:57:22Z)

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.