Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:2501.19307

Quantum-Inspired Fidelity-based Divergence

Yifeng Peng, Dantong Li, Xinyi Li, Zhiding Liang, Yongshan Ding, Ying Wang

Published Jan 31, 2025• 0 citations
arXivHTMLPDF

Abstract

Kullback--Leibler (KL) divergence is a fundamental measure of the dissimilarity between two probability distributions, but it can become unstable in high-dimensional settings due to its sensitivity to mismatches in distributional support. To address robustness limitations, we propose a novel Quantum-Inspired Fidelity-based Divergence (QIF), leveraging quantum information principles yet efficiently computable on classical hardware. Compared to KL divergence, QIF demonstrates improved numerical stability under partial or near-disjoint support conditions, thereby reducing the need for extensive regularization in specific scenarios. Moreover, QIF admits well-defined theoretical bounds and continuous similarity measures. Building on this, we introduce a novel regularization method, QR-Drop, which utilizes QIF to improve generalization in machine learning models. Empirical results show that QR-Drop effectively mitigates overfitting and outperforms state-of-the-art methods.

DOI
10.48550/arxiv.2501.19307

Related papers

Kullback-Leibler divergence between quantum distributions, and its upper-bound

Vincenzo Bonnici

This work presents an upper-bound to value that the Kullback-Leibler (KL) divergence can reach for a class of probability distributions called quantum distributions (QD). The aim is to find a distribution $U$ which maximizes the KL divergence from a given distribution $P$ under the assumption that $P$ and $U$ have been generated by distributing a given discrete quantity, a quantum. Quantum distributions naturally represent a wide range of probability distributions that are used in practical applications. Moreover, such a class of distributions can be obtained as an approximation of any probability distribution. The retrieving of an upper-bound for the entropic divergence is here shown to be possible under the condition that the compared distributions are quantum distributions over the same quantum value, thus they become comparable. Thus, entropic divergence acquires a more powerful meaning when it is applied to comparable distributions. This aspect should be taken into account in future developments of divergences. The theoretical findings are used for proposing a notion of normalized KL divergence that is empirically shown to behave differently from already known measures.

arXiv:2008.05932•Aug 13, 2020•3 citations

A maximum value for the Kullback-Leibler divergence between quantum discrete distributions

Vincenzo Bonnici

This work presents an upper-bound for the maximum value that the Kullback-Leibler (KL) divergence can reach for a newly introduced class of probability distributions, called quantum distributions (QD). In particular, the aim is to find a discrete distribution $U$ which maximize the finite KL divergence from a given $P$ under the assumption that $P$ and $U$ have been generated by distributing a fixed discretized quantity. The theoretical findings are used for proposing a notion of normalized KL divergence that is empirically shown to behave differently from already known measures.

•Aug 13, 2020•0 citations

Quantum-Inspired Probability Distributions for Machine Learning

Jincheng Zhang

Quantum-inspired probability distributions are proposed as a novel approach to enhance the robustness and accuracy of machine learning algorithms. This paper explores the potential of leveraging quantum-like phenomena, specifically superposition and entanglement, to mitigate the effects of noise and overfitting, resulting in improved model performance across a range of applications. We detail the core mechanism, outlining the design and implementation of a new distribution family centered around these principles. This work presents a theoretical framework and preliminary experimental results demonstrating the effectiveness of this approach in several machine learning tasks. The core claim is that these distributions offer a significant advancement in robustness compared to traditional probability distributions.

•Sep 2, 2026•0 citations

Quantum-Inspired Regularization for Machine Learning

Jincheng Zhang

Quantum-inspired regularization is a novel approach to machine learning that leverages the principles of quantum error correction to enhance model robustness and accuracy. This paper explores the integration of quantum-inspired error correction mechanisms into the optimization process, aiming to mitigate the sensitivity of models to noise and improve generalization performance. We present a theoretical framework outlining the key components of this method and demonstrate its effectiveness through a series of simulations, showcasing a significant reduction in error sensitivity compared to traditional regularization techniques.

•Aug 31, 2026•0 citations

Nonadditive generalization of the quantum Kullback-Leibler divergence for measuring the degree of purification

Sumiyoshi Abe

The Kullback-Leibler divergence offers an information-theoretic basis for measuring the difference between two given distributions. Its quantum analog, however, fails to play a corresponding role for comparing two density matrices, if the reference states are pure states. Here it is shown that nonadditive quantum information theory inspired by nonextensive statistical mechanics is free from such a difficulty and the associated quantity, termed the quantum q-divergence, can in fact be a good information-theoretic measure of the degree of state purification. The correspondence relation between the ordinary divergence and the q-divergence is violated for the pure reference states, in general.

arXiv:quant-ph/0301136•Physical Review A•Sep 10, 2003•63 citations

Noise-Robust Quantum Generative Models for Distribution Learning and Efficient Data Loading

Ankit Kumar Mandusia

Abstract Quantum machine learning offers powerful advantages in processing high-dimensional data, but loading classical probability distributions into quantum states remains a major challenge. This work extends quantum generative adversarial networks (qGANs) to overcome two critical limitations of early approaches: poor robustness to noise on near-term devices and limited capacity to model complex, multi-modal distributions. We introduce hybrid qGAN architectures that combine Wasserstein GAN with gradient penalty (WGAN-GP) and maximum mean discrepancy (MMD) losses with expressive quantum circuits, including quantum convolutional neural networks (QCNNs) and EfficientSU2 ansätze. Training is performed using seamless PyTorch–Qiskit integration, enabling stable optimization on both simulators and real quantum hardware.Experiments on 2D Gaussian and log-normal distributions—directly relevant to financial modeling—show faster convergence and higher fidelity than prior qGAN designs, with up to 80% lower Wasserstein distance under 5% depolarizing noise. The framework is further validated through European call option pricing via quantum amplitude estimation (QAE), achieving pricing errors below 1% on IBM’s 20-qubit superconducting systems. Theoretical analysis confirms gradient stability under noise, while empirical results include comprehensive noise sweeps, t-SNE latent space visualizations, and rigorous statistical tests (KL divergence, KS tests). This work establishes a practical path toward quantum advantage in generative modeling, with fully open-source code for reproducibility.

•Nov 18, 2025•0 citations
Kullback--Leibler (KL) divergence is a fundamental measure of the dissimilarity between two probability distributions, but it can become unstable in high-dimensional settings due to its sensitivity to mismatches in distributional support. To address robustness limitations, we propose a novel Quantum-Inspired Fidelity-based Divergence (QIF), leveraging quantum information principles yet efficiently computable on classical hardware. Compared to KL divergence, QIF demonstrates improved numerical stability under partial or near-disjoint support conditions, thereby reducing the need for extensive regularization in specific scenarios. Moreover, QIF admits well-defined theoretical bounds and continuous similarity measures. Building on this, we introduce a novel regularization method, QR-Drop, which utilizes QIF to improve generalization in machine learning models. Empirical results show that QR-Drop effectively mitigates overfitting and outperforms state-of-the-art methods.
This work presents an upper-bound to value that the Kullback-Leibler (KL) divergence can reach for a class of probability distributions called quantum distributions (QD). The aim is to find a distribution UUU which maximizes the KL divergence from a given distribution PPP under the assumption that PPP and UUU have been generated by distributing a given discrete quantity, a quantum. Quantum distributions naturally represent a wide range of probability distributions that are used in practical applications. Moreover, such a class of distributions can be obtained as an approximation of any probability distribution. The retrieving of an upper-bound for the entropic divergence is here shown to be possible under the condition that the compared distributions are quantum distributions over the same quantum value, thus they become comparable. Thus, entropic divergence acquires a more powerful meaning when it is applied to comparable distributions. This aspect should be taken into account in future developments of divergences. The theoretical findings are used for proposing a notion of normalized KL divergence that is empirically shown to behave differently from already known measures.
This work presents an upper-bound for the maximum value that the Kullback-Leibler (KL) divergence can reach for a newly introduced class of probability distributions, called quantum distributions (QD). In particular, the aim is to find a discrete distribution UUU which maximize the finite KL divergence from a given PPP under the assumption that PPP and UUU have been generated by distributing a fixed discretized quantity. The theoretical findings are used for proposing a notion of normalized KL divergence that is empirically shown to behave differently from already known measures.
Quantum-inspired probability distributions are proposed as a novel approach to enhance the robustness and accuracy of machine learning algorithms. This paper explores the potential of leveraging quantum-like phenomena, specifically superposition and entanglement, to mitigate the effects of noise and overfitting, resulting in improved model performance across a range of applications. We detail the core mechanism, outlining the design and implementation of a new distribution family centered around these principles. This work presents a theoretical framework and preliminary experimental results demonstrating the effectiveness of this approach in several machine learning tasks. The core claim is that these distributions offer a significant advancement in robustness compared to traditional probability distributions.
Quantum-inspired regularization is a novel approach to machine learning that leverages the principles of quantum error correction to enhance model robustness and accuracy. This paper explores the integration of quantum-inspired error correction mechanisms into the optimization process, aiming to mitigate the sensitivity of models to noise and improve generalization performance. We present a theoretical framework outlining the key components of this method and demonstrate its effectiveness through a series of simulations, showcasing a significant reduction in error sensitivity compared to traditional regularization techniques.
The Kullback-Leibler divergence offers an information-theoretic basis for measuring the difference between two given distributions. Its quantum analog, however, fails to play a corresponding role for comparing two density matrices, if the reference states are pure states. Here it is shown that nonadditive quantum information theory inspired by nonextensive statistical mechanics is free from such a difficulty and the associated quantity, termed the quantum q-divergence, can in fact be a good information-theoretic measure of the degree of state purification. The correspondence relation between the ordinary divergence and the q-divergence is violated for the pure reference states, in general.
Abstract Quantum machine learning offers powerful advantages in processing high-dimensional data, but loading classical probability distributions into quantum states remains a major challenge. This work extends quantum generative adversarial networks (qGANs) to overcome two critical limitations of early approaches: poor robustness to noise on near-term devices and limited capacity to model complex, multi-modal distributions. We introduce hybrid qGAN architectures that combine Wasserstein GAN with gradient penalty (WGAN-GP) and maximum mean discrepancy (MMD) losses with expressive quantum circuits, including quantum convolutional neural networks (QCNNs) and EfficientSU2 ansätze. Training is performed using seamless PyTorch–Qiskit integration, enabling stable optimization on both simulators and real quantum hardware.Experiments on 2D Gaussian and log-normal distributions—directly relevant to financial modeling—show faster convergence and higher fidelity than prior qGAN designs, with up to 80% lower Wasserstein distance under 5% depolarizing noise. The framework is further validated through European call option pricing via quantum amplitude estimation (QAE), achieving pricing errors below 1% on IBM’s 20-qubit superconducting systems. Theoretical analysis confirms gradient stability under noise, while empirical results include comprehensive noise sweeps, t-SNE latent space visualizations, and rigorous statistical tests (KL divergence, KS tests). This work establishes a practical path toward quantum advantage in generative modeling, with fully open-source code for reproducibility.