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:2611.16903

Uncertainty-aware quantum machine learning based on Dempster-Shafer theory

Marie Kempkes, Jakob Spiegelberg, Evert van Nieuwenburg, Vedran Dunjko

Published Nov 12, 2026• 0 citations
arXivHTMLPDF

Abstract

A large body of work in quantum machine learning has centered on characterizing the expressivity and performance of quantum models. In contrast, little attention has been paid to the reliability of predictions beyond average performance measures such as accuracy. In this work, we introduce an uncertainty-aware quantum classification framework based on Dempster-Shafer theory. This theory allows to express uncertainty through weights assigned not only to individual classification labels but to sets of labels. We introduce a training strategy that enables parameterized quantum circuits to learn such assignments directly from data. This provides an efficient way to prepare the corresponding quantum states and enables the application of different post-processing strategies. In particular, it offers a practical setting for quantum Dempster-Shafer fusion procedures with polynomial resource requirements, even when the corresponding mass functions are defined over exponentially many subsets. Overall, our work represents a step toward enabling quantum models to express uncertainty about their predictions, thereby extending the evaluation of quantum models beyond predictive performance alone.

DOI
10.48550/arxiv.2611.16903

Related papers

Bayesian Variational Quantum Classifiers for Uncertainty-Aware Quantum Machine Learning

Bhavna Saini, Vivek Kumar Verma, Nidhi Kushwaha +1 more

Quantum machine learning is increasingly being explored as a way to tackle complex data analysis problems by combining quantum computing with modern learning techniques. In particular, variational quantum classifiers have attracted considerable attention because they can be implemented on current noisy intermediate-scale quantum (NISQ) devices. Despite their potential, most existing variational quantum classifiers rely on deterministic optimization of circuit parameters. Such approaches can struggle when the available training data are limited or noisy, often leading to unstable predictions and reduced reliability. To address these challenges, recent research has begun incorporating Bayesian learning principles into variational quantum models. By treating circuit parameters as probabilistic variables rather than fixed values, Bayesian variational quantum classifiers allow uncertainty to be explicitly modeled during the learning process. This makes it possible to obtain predictions that are not only accurate but also accompanied by meaningful confidence estimates. In this paper, we present a detailed analysis of Bayesian variational quantum classifiers and their role in enabling uncertainty-aware quantum machine learning. The discussion covers the foundations of variational quantum circuits, Bayesian parameter estimation, and probabilistic prediction mechanisms within hybrid quantum-classical learning frameworks. Different training and inference strategies, including variational Bayesian optimization and sampling-based methods, are examined to highlight their effectiveness in capturing parameter uncertainty in quantum circuits. We also discuss emerging application areas, such as healthcare analytics, financial forecasting, and scientific data analysis, where uncertainty-aware quantum learning models can provide valuable insights.

International Conference on Data Science for Cyber-Physical Systems Resilience using Advanced Applications (ICDCA)•Jun 1, 2026•0 citations

Uncertainty-Aware Variational Quantum Feature Learning for Reliable High-Dimensional Classification

Syed Basha Shaik, Srihari Varma Mantena, Shaik Janbhasha +4 more

Abstract High-dimensional classification remains challenging for hybrid quantum–classical learning because conventional variational quantum feature learning methods often assign equal importance to all input features and do not explicitly account for feature uncertainty. This paper presents Uncertainty-Aware Variational Quantum Feature Learning (UVQFL) , a hybrid framework that combines adaptive feature weighting, uncertainty estimation, and variational quantum representation learning for reliable classification. The proposed framework incorporates an Adaptive Feature Confidence Module (AFCM) to estimate the contribution of individual input features before quantum encoding and an Uncertainty Estimation Module (UEM) to characterize feature reliability during the learning process. The confidence-weighted features are encoded using a variational quantum circuit and subsequently processed by an enhanced multi-layer classical classifier. Benchmark datasets, including MNIST, Fashion-MNIST, and PathMNIST, are employed during data preprocessing and quantum feature generation to establish the proposed framework. The detailed experimental evaluation, performance analysis, and comparative results presented in this study are conducted on the MNIST benchmark dataset to provide a comprehensive assessment of the proposed architecture. A systematic investigation using 4-, 6-, and 8-qubit hybrid models examines the influence of quantum feature dimensionality on classification performance. The final UVQFL model achieves a best validation accuracy of 84.62% , with 84.60% precision , 84.62% recall , and an 84.59% F1-score . The results indicate that integrating adaptive feature confidence with uncertainty-aware quantum feature learning improves feature representation quality and enhances the reliability of hybrid quantum–classical classification.

•Aug 14, 2026•0 citations

An Uncertainty-Aware Adaptive Hybrid Quantum-Classical Learning Framework for Drug Discovery

A. Santhosh Nantha, M. Jayalakshmi, K. Maharajan +2 more

Quantum machine learning (QML) has attracted significant interest in drug discovery due to its potential to model complex molecular interactions and improve prediction accuracy for protein-ligand binding affinity. However, most existing hybrid quantum-classical approaches rely on static architectures that apply quantum computation uniformly across all samples, regardless of molecular complexity or prediction confidence. Such static execution strategies lead to inefficient quantum resource utilization and limit scalability on near-term noisy intermediatescale quantum (NISQ) devices. In this paper, we propose an uncertainty-aware adaptive hybrid quantum-classical learning framework for protein-ligand binding affinity prediction that treats quantum computation as a selectively invoked resource rather than a constant component of the learning pipeline. The proposed framework first employs a classical ensemble-based regression model to generate binding affinity predictions along with an associated uncertainty estimate derived from ensemble variance. This uncertainty information is then used by a decision mechanism to dynamically select between classical-only inference, shallow quantum circuit execution, or deep quantum circuit execution. By adapting the quantum circuit depth and invocation frequency to the predicted uncertainty, the framework enables decision-centric allocation of quantum resources. Experimental evaluation of binding affinity data demonstrates that the recommended adaptive framework utilizes around 76 % less quantum resources than static hybrid quantum learning methods that use fixed-depth quantum circuits for all samples. The framework also preserves decision-level functionality and a steady inference workflow. These findings highlight the effectiveness of uncertainty-guided adaptivity as a practical strategy and offer a scalable foundation for future quantum-enhanced molecular learning.method for incorporating QML into drug development pipelines while taking hardware constraints into consideration.

International Conference on Signal, Systems, and Computing for Next-Gen Automation (ICSSCNA)•May 4, 2026•0 citations

Quantum ensembles of quantum classifiers

Maria Schuld, Francesco Petruccione

Quantum machine learning witnesses an increasing amount of quantum algorithms for data-driven decision making, a problem with potential applications ranging from automated image recognition to medical diagnosis. Many of those algorithms are implementations of quantum classifiers, or models for the classification of data inputs with a quantum computer. Following the success of collective decision making with ensembles in classical machine learning, this paper introduces the concept of quantum ensembles of quantum classifiers. Creating the ensemble corresponds to a state preparation routine, after which the quantum classifiers are evaluated in parallel and their combined decision is accessed by a single-qubit measurement. This framework naturally allows for exponentially large ensembles in which - similar to Bayesian learning - the individual classifiers do not have to be trained. As an example, we analyse an exponentially large quantum ensemble in which each classifier is weighed according to its performance in classifying the training data, leading to new results for quantum as well as classical machine learning.

arXiv:1704.02146•Scientific Reports•Feb 5, 2018•93 citations

Quantum Machine Learning: A Theoretical Framework for Generalization

Jincheng Zhang

This paper delves into the nascent field of quantum machine learning (QML) by establishing a theoretical framework for understanding and predicting the generalization capabilities of QML algorithms. Currently, the practical success of QML algorithms is often attributed to empirical observation, lacking a robust theoretical underpinning. This work addresses this gap by leveraging concepts from quantum information theory to analyze the generalization performance of various QML algorithms. We propose a model that connects algorithm parameters, dataset characteristics, and the resulting generalization error, expressed mathematically as: Γ(θ, D) = ED[|f(x) - ψ(x)|] where: Γ(θ, D) represents the generalization error; θ represents the algorithm's parameters; D represents the dataset; ED[ ] denotes the expected value with respect to the dataset D; f(x) is the model's prediction; ψ(x) is the true label or target. Furthermore, we explore the potential for quantum advantage in generalization, considering the impact of quantum entanglement and superposition on reducing the risk of overfitting. The core contribution lies in providing a methodological approach for quantifying and improving generalization performance in QML, moving beyond purely experimental evaluations. We introduce a novel metric, the Quantum Generalization Index (QGI), defined as: QGI = (1 - Γ(θ, D)) / (1 + Γ(θ, D)) This index provides a normalized measure of generalization performance, allowing for comparative analysis of different QML algorithms and parameter settings. Ultimately, this research aims to provide a foundational understanding for the design and development of more reliable and robust quantum machine learning systems.

•Aug 29, 2026•0 citations

Quantum Information Fusion and Correction under the Transferable Belief Model

Qianli Zhou, Hao Luo, Lipeng Pan +2 more

The transferable belief model (TBM), developed within Dempster-Shafer theory, represents ambiguity and partial ignorance through set-valued belief masses. Its classical operations, however, can grow combinatorially over the power set. We formulate TBM reasoning on quantum circuits using the mass function quantum state (MFQS). The resulting framework covers belief representation, credal-level fusion and correction, product-space operations, and the generalized Bayesian theorem. For prepared MFQS inputs, CCR and the entire $α$-junction use element-wise gate modules whose logical counts are linear in the number of frame elements. These circuits avoid explicit operation-stage updates over all $2^n$ focal-set coordinates. A noisy multi-domain classifier-fusion example further demonstrates interpretable source fusion and improved robustness to gate-control noise. The results establish Dempster-Shafer structures as a semantic layer for quantum information processing beyond singleton-probability representations.

arXiv:2410.08949•Oct 11, 2024•2 citations
A large body of work in quantum machine learning has centered on characterizing the expressivity and performance of quantum models. In contrast, little attention has been paid to the reliability of predictions beyond average performance measures such as accuracy. In this work, we introduce an uncertainty-aware quantum classification framework based on Dempster-Shafer theory. This theory allows to express uncertainty through weights assigned not only to individual classification labels but to sets of labels. We introduce a training strategy that enables parameterized quantum circuits to learn such assignments directly from data. This provides an efficient way to prepare the corresponding quantum states and enables the application of different post-processing strategies. In particular, it offers a practical setting for quantum Dempster-Shafer fusion procedures with polynomial resource requirements, even when the corresponding mass functions are defined over exponentially many subsets. Overall, our work represents a step toward enabling quantum models to express uncertainty about their predictions, thereby extending the evaluation of quantum models beyond predictive performance alone.
Quantum machine learning is increasingly being explored as a way to tackle complex data analysis problems by combining quantum computing with modern learning techniques. In particular, variational quantum classifiers have attracted considerable attention because they can be implemented on current noisy intermediate-scale quantum (NISQ) devices. Despite their potential, most existing variational quantum classifiers rely on deterministic optimization of circuit parameters. Such approaches can struggle when the available training data are limited or noisy, often leading to unstable predictions and reduced reliability. To address these challenges, recent research has begun incorporating Bayesian learning principles into variational quantum models. By treating circuit parameters as probabilistic variables rather than fixed values, Bayesian variational quantum classifiers allow uncertainty to be explicitly modeled during the learning process. This makes it possible to obtain predictions that are not only accurate but also accompanied by meaningful confidence estimates. In this paper, we present a detailed analysis of Bayesian variational quantum classifiers and their role in enabling uncertainty-aware quantum machine learning. The discussion covers the foundations of variational quantum circuits, Bayesian parameter estimation, and probabilistic prediction mechanisms within hybrid quantum-classical learning frameworks. Different training and inference strategies, including variational Bayesian optimization and sampling-based methods, are examined to highlight their effectiveness in capturing parameter uncertainty in quantum circuits. We also discuss emerging application areas, such as healthcare analytics, financial forecasting, and scientific data analysis, where uncertainty-aware quantum learning models can provide valuable insights.
Abstract High-dimensional classification remains challenging for hybrid quantum–classical learning because conventional variational quantum feature learning methods often assign equal importance to all input features and do not explicitly account for feature uncertainty. This paper presents Uncertainty-Aware Variational Quantum Feature Learning (UVQFL) , a hybrid framework that combines adaptive feature weighting, uncertainty estimation, and variational quantum representation learning for reliable classification. The proposed framework incorporates an Adaptive Feature Confidence Module (AFCM) to estimate the contribution of individual input features before quantum encoding and an Uncertainty Estimation Module (UEM) to characterize feature reliability during the learning process. The confidence-weighted features are encoded using a variational quantum circuit and subsequently processed by an enhanced multi-layer classical classifier. Benchmark datasets, including MNIST, Fashion-MNIST, and PathMNIST, are employed during data preprocessing and quantum feature generation to establish the proposed framework. The detailed experimental evaluation, performance analysis, and comparative results presented in this study are conducted on the MNIST benchmark dataset to provide a comprehensive assessment of the proposed architecture. A systematic investigation using 4-, 6-, and 8-qubit hybrid models examines the influence of quantum feature dimensionality on classification performance. The final UVQFL model achieves a best validation accuracy of 84.62% , with 84.60% precision , 84.62% recall , and an 84.59% F1-score . The results indicate that integrating adaptive feature confidence with uncertainty-aware quantum feature learning improves feature representation quality and enhances the reliability of hybrid quantum–classical classification.
Quantum machine learning (QML) has attracted significant interest in drug discovery due to its potential to model complex molecular interactions and improve prediction accuracy for protein-ligand binding affinity. However, most existing hybrid quantum-classical approaches rely on static architectures that apply quantum computation uniformly across all samples, regardless of molecular complexity or prediction confidence. Such static execution strategies lead to inefficient quantum resource utilization and limit scalability on near-term noisy intermediatescale quantum (NISQ) devices. In this paper, we propose an uncertainty-aware adaptive hybrid quantum-classical learning framework for protein-ligand binding affinity prediction that treats quantum computation as a selectively invoked resource rather than a constant component of the learning pipeline. The proposed framework first employs a classical ensemble-based regression model to generate binding affinity predictions along with an associated uncertainty estimate derived from ensemble variance. This uncertainty information is then used by a decision mechanism to dynamically select between classical-only inference, shallow quantum circuit execution, or deep quantum circuit execution. By adapting the quantum circuit depth and invocation frequency to the predicted uncertainty, the framework enables decision-centric allocation of quantum resources. Experimental evaluation of binding affinity data demonstrates that the recommended adaptive framework utilizes around 76 % less quantum resources than static hybrid quantum learning methods that use fixed-depth quantum circuits for all samples. The framework also preserves decision-level functionality and a steady inference workflow. These findings highlight the effectiveness of uncertainty-guided adaptivity as a practical strategy and offer a scalable foundation for future quantum-enhanced molecular learning.method for incorporating QML into drug development pipelines while taking hardware constraints into consideration.
Quantum machine learning witnesses an increasing amount of quantum algorithms for data-driven decision making, a problem with potential applications ranging from automated image recognition to medical diagnosis. Many of those algorithms are implementations of quantum classifiers, or models for the classification of data inputs with a quantum computer. Following the success of collective decision making with ensembles in classical machine learning, this paper introduces the concept of quantum ensembles of quantum classifiers. Creating the ensemble corresponds to a state preparation routine, after which the quantum classifiers are evaluated in parallel and their combined decision is accessed by a single-qubit measurement. This framework naturally allows for exponentially large ensembles in which - similar to Bayesian learning - the individual classifiers do not have to be trained. As an example, we analyse an exponentially large quantum ensemble in which each classifier is weighed according to its performance in classifying the training data, leading to new results for quantum as well as classical machine learning.
This paper delves into the nascent field of quantum machine learning (QML) by establishing a theoretical framework for understanding and predicting the generalization capabilities of QML algorithms. Currently, the practical success of QML algorithms is often attributed to empirical observation, lacking a robust theoretical underpinning. This work addresses this gap by leveraging concepts from quantum information theory to analyze the generalization performance of various QML algorithms. We propose a model that connects algorithm parameters, dataset characteristics, and the resulting generalization error, expressed mathematically as: Γ(θ, D) = ED[|f(x) - ψ(x)|] where: Γ(θ, D) represents the generalization error; θ represents the algorithm's parameters; D represents the dataset; ED[ ] denotes the expected value with respect to the dataset D; f(x) is the model's prediction; ψ(x) is the true label or target. Furthermore, we explore the potential for quantum advantage in generalization, considering the impact of quantum entanglement and superposition on reducing the risk of overfitting. The core contribution lies in providing a methodological approach for quantifying and improving generalization performance in QML, moving beyond purely experimental evaluations. We introduce a novel metric, the Quantum Generalization Index (QGI), defined as: QGI = (1 - Γ(θ, D)) / (1 + Γ(θ, D)) This index provides a normalized measure of generalization performance, allowing for comparative analysis of different QML algorithms and parameter settings. Ultimately, this research aims to provide a foundational understanding for the design and development of more reliable and robust quantum machine learning systems.
The transferable belief model (TBM), developed within Dempster-Shafer theory, represents ambiguity and partial ignorance through set-valued belief masses. Its classical operations, however, can grow combinatorially over the power set. We formulate TBM reasoning on quantum circuits using the mass function quantum state (MFQS). The resulting framework covers belief representation, credal-level fusion and correction, product-space operations, and the generalized Bayesian theorem. For prepared MFQS inputs, CCR and the entire ααα-junction use element-wise gate modules whose logical counts are linear in the number of frame elements. These circuits avoid explicit operation-stage updates over all 2n2^n2n focal-set coordinates. A noisy multi-domain classifier-fusion example further demonstrates interpretable source fusion and improved robustness to gate-control noise. The results establish Dempster-Shafer structures as a semantic layer for quantum information processing beyond singleton-probability representations.