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:2509.12465 · Physical Review A

Efficient privacy-preserving training of quantum neural networks through ensemble encoding of quantum states

Gaoyuan Wang, Jonathan Warrell, Mark Gerstein

Published Jul 6, 2026• 1 citation
arXivHTMLPDF

Abstract

Quantum computing is gaining popularity due to its potential to detect complex patterns in data by leveraging unique quantum phenomena. It is particularly promising for complex data applications, such as those in biomedicine. However, in these contexts, it is often necessary to share and then aggregate data from multiple participants to achieve sufficient statistical power. Sharing data, especially sensitive information such as personal medical records, raises significant privacy concerns. To address these challenges, we propose a quantum-native method for encoding entire data cohorts directly into special quantum states, which we refer to as composite states. These quantum states provide a generic encoding suitable for a wide variety of downstream computations while preventing the inference of individual-level information. Quantum computations can be performed directly on composite states without accessing the underlying raw data. Building on this, we present the theoretical foundations of our scheme's utility and privacy guarantees, demonstrating resistance to membership inference attacks as measured via differential privacy. Furthermore, we introduce protocols that support multiparty collaborative quantum neural network training across diverse domains. Finally, we validate the effectiveness of composite states on three different datasets, focusing on genomic studies while also indicating how the approach can be applied in other domains without adaptation.

DOI
10.1103/81xl-nr9q

Related papers

Quantum-Enhanced Differential Privacy with Entangled Data Sharing

Jincheng Zhang

This paper explores a novel approach to achieving differential privacy in quantum computations through entangled data sharing. Traditional differential privacy methods struggle to scale effectively in complex quantum algorithms, largely due to the inherent sensitivity of quantum states. We propose encoding sensitive data into entangled quantum states and performing computations on these states. The utilization of entanglement fundamentally alters the privacy landscape, offering the potential for enhanced privacy guarantees compared to classical differential privacy techniques. The core idea is that the correlations introduced by entanglement will make it significantly harder to infer information about individual data points from the resulting quantum state, even with access to a significant amount of quantum computation. We provide a theoretical framework outlining this approach and discuss the potential implications for secure quantum data analysis. The analysis focuses on the impact of entanglement on the privacy loss parameter, δ, demonstrating a possible reduction in privacy loss compared to standard methods. This work represents a crucial step towards realizing the full potential of quantum computing while simultaneously maintaining strong privacy protections.

•Sep 5, 2026•0 citations

Hybrid Quantum-Classical Computing for Privacy-Preserving Machine Learning on Edge Devices

B. Dinesh, M. Thangamani, B. Paulchamy +3 more

The edge computing environment faces a challenge in maintaining data privacy while performing machine learning tasks. The current solutions for privacy preservation are inefficient or do not provide security guarantees, causing a critical gap in the edge computing environment, which is resource-constrained. The aim of this research work is to design a hybrid quantum-classical computing system that addresses the challenges of privacy preservation and efficiency in machine learning tasks. We design a novel architecture combining variational quantum circuits for secure data encoding with classical neural networks for scalable inference, implemented through federated learning protocols enhanced with quantum differential privacy mechanisms. Experimental evaluation demonstrates a 35% reduction in communication overhead compared to traditional federated learning approaches while maintaining 99.2% classification accuracy across diverse datasets. The quantum-enhanced privacy mechanisms achieve provable security guarantees with information-theoretic bounds, outperforming conventional cryptographic methods by 28% in computational overhead reduction. This hybrid approach establishes a new paradigm for secure and efficient machine learning deployment on edge devices, enabling privacy-preserving artificial intelligence in healthcare, finance, and IoT applications.

IEEE International Conference on Emerging Computing and Intelligent Technologies (ICoECIT)•Jan 28, 2026•0 citations

Quantum neural network classifier with differential privacy

Guodong Li, Fangce Yu, Qingle Wang +3 more

Abstract Quantum neural networks (QNN) hold the potential for quantum advantages but are confronted with privacy leakage risks that could undermine users’ benefits. To address this issue, we propose an enhanced QNN model integrated with differential privacy. After converting classical data into quantum states using encoding circuit, we apply data re-uploading technology to embed the encoding circuit multiple times into the overall circuit. The classification outcomes are derived by measuring the final quantum state. During the optimization process of QNN, we introduce Gaussian noise into the gradient of the loss function to ensure differential privacy. Our theoretical analysis demonstrates that in differential privacy settings, the model effectively protects privacy while demonstrating good performance. Our experimental analysis demonstrates that on four standard datasets, our model achieves the same accuracy as classical neural network (NN) while utilizing only 22% of the parameters of NN, and exhibiting lower performance loss compared to other QNN.

Physica Scripta•Feb 4, 2025•2 citations

Quantum Entanglement-Based Differential Privacy

Jincheng Zhang

This paper explores a novel approach to differential privacy utilizing quantum entanglement. Traditional differential privacy relies on adding noise to queries to obscure individual data contributions, a technique often limited by the dimensionality of the data and the potential for sophisticated attacks. We propose encoding data points into entangled qubit states, thereby leveraging the inherent correlations provided by entanglement to achieve stronger privacy guarantees. The core idea involves performing computations on these entangled states, minimizing the need for direct data exposure. Crucially, we introduce a mechanism for carefully calibrating measurement noise, exploiting the unique properties of entanglement to control the privacy loss. Our framework demonstrates the potential to significantly improve privacy levels compared to classical methods, especially when dealing with high-dimensional datasets, offering a new paradigm for privacy-preserving data analysis in the quantum realm. The theoretical analysis presents a novel framework for quantifying privacy loss, considering both the entanglement resource requirements and the calibrated noise levels. This work lays the foundation for future research into quantum-enhanced differential privacy and its implications for secure data processing.

•Sep 5, 2026•0 citations

Federated Quantum Machine Learning for Secure Multi-Party Genomic Data Analysis: A Hybrid Architecture with Differential Privacy

David Jumpa Malem Sembiring, Sinek Mehuli Br Perangin Angin

The exponential growth of genomic data has created opportunities for collaborative biomedical research while posing significant privacy and security challenges. Traditional centralized machine learning approaches risk data breaches and regulatory non-compliance. This study proposes a novel hybrid architecture combining Federated Quantum Machine Learning (FQML) with Differential Privacy (DP) for secure multi-party genomic data analysis. Our simulated framework enables institutions to collaboratively train quantum machine learning models on distributed genomic datasets without exposing sensitive data. Using variational quantum circuits within a federated learning framework, we implement privacy-preserving models across three simulated genomic data silos. Gaussian noise injection via DP mechanisms during gradient updates ensures formal privacy guarantees (ε = 0.85). Experimental results on synthetic datasets demonstrate that FQML achieves 93.4% predictive accuracy in genome-wide association studies simulation, outperforming classical federated models in convergence speed by 27% and reducing communication overhead by 19%. These findings suggest FQML with DP provides a promising foundation for future secure genomic collaboration, though real-world validation remains necessary.

Journal of Novel Engineering Science and Technology•May 4, 2026•0 citations

Multiparty private summation protocol based on two-state quantum-mechanical system

Jason Lin, Shao-Lun Huang, Chun‐Wei Yang +1 more

With the rapid advancement of information technology, data sharing has become increasingly accessible, leading to a heightened need for robust personal data protection. One important application in privacy-preserving computing is the aggregation of information when collaboratively establishing AI models through public distributed networks. To counter the threat posed by quantum computing to encrypted data, various quantum private summation (QPS) protocols have been proposed thus far. However, some of these existing protocols operate solely under modulo 2, while other approaches for modulo d often rely on impractically high-dimensional qudits. Therefore, this study proposes an innovative multiparty QPS method that balances participant data sharing and privacy without requiring high-dimensional photons. The proposed QPS protocol enables participants to contribute aggregated information to third parties without disclosing individual data. A security analysis further demonstrates that the proposed QPS effectively counters common eavesdropping attacks, ensuring reliable protection of personal data.

Scientific Reports•Jul 2, 2025•2 citations
Quantum computing is gaining popularity due to its potential to detect complex patterns in data by leveraging unique quantum phenomena. It is particularly promising for complex data applications, such as those in biomedicine. However, in these contexts, it is often necessary to share and then aggregate data from multiple participants to achieve sufficient statistical power. Sharing data, especially sensitive information such as personal medical records, raises significant privacy concerns. To address these challenges, we propose a quantum-native method for encoding entire data cohorts directly into special quantum states, which we refer to as composite states. These quantum states provide a generic encoding suitable for a wide variety of downstream computations while preventing the inference of individual-level information. Quantum computations can be performed directly on composite states without accessing the underlying raw data. Building on this, we present the theoretical foundations of our scheme's utility and privacy guarantees, demonstrating resistance to membership inference attacks as measured via differential privacy. Furthermore, we introduce protocols that support multiparty collaborative quantum neural network training across diverse domains. Finally, we validate the effectiveness of composite states on three different datasets, focusing on genomic studies while also indicating how the approach can be applied in other domains without adaptation.
This paper explores a novel approach to achieving differential privacy in quantum computations through entangled data sharing. Traditional differential privacy methods struggle to scale effectively in complex quantum algorithms, largely due to the inherent sensitivity of quantum states. We propose encoding sensitive data into entangled quantum states and performing computations on these states. The utilization of entanglement fundamentally alters the privacy landscape, offering the potential for enhanced privacy guarantees compared to classical differential privacy techniques. The core idea is that the correlations introduced by entanglement will make it significantly harder to infer information about individual data points from the resulting quantum state, even with access to a significant amount of quantum computation. We provide a theoretical framework outlining this approach and discuss the potential implications for secure quantum data analysis. The analysis focuses on the impact of entanglement on the privacy loss parameter, δ, demonstrating a possible reduction in privacy loss compared to standard methods. This work represents a crucial step towards realizing the full potential of quantum computing while simultaneously maintaining strong privacy protections.
The edge computing environment faces a challenge in maintaining data privacy while performing machine learning tasks. The current solutions for privacy preservation are inefficient or do not provide security guarantees, causing a critical gap in the edge computing environment, which is resource-constrained. The aim of this research work is to design a hybrid quantum-classical computing system that addresses the challenges of privacy preservation and efficiency in machine learning tasks. We design a novel architecture combining variational quantum circuits for secure data encoding with classical neural networks for scalable inference, implemented through federated learning protocols enhanced with quantum differential privacy mechanisms. Experimental evaluation demonstrates a 35% reduction in communication overhead compared to traditional federated learning approaches while maintaining 99.2% classification accuracy across diverse datasets. The quantum-enhanced privacy mechanisms achieve provable security guarantees with information-theoretic bounds, outperforming conventional cryptographic methods by 28% in computational overhead reduction. This hybrid approach establishes a new paradigm for secure and efficient machine learning deployment on edge devices, enabling privacy-preserving artificial intelligence in healthcare, finance, and IoT applications.
Abstract Quantum neural networks (QNN) hold the potential for quantum advantages but are confronted with privacy leakage risks that could undermine users’ benefits. To address this issue, we propose an enhanced QNN model integrated with differential privacy. After converting classical data into quantum states using encoding circuit, we apply data re-uploading technology to embed the encoding circuit multiple times into the overall circuit. The classification outcomes are derived by measuring the final quantum state. During the optimization process of QNN, we introduce Gaussian noise into the gradient of the loss function to ensure differential privacy. Our theoretical analysis demonstrates that in differential privacy settings, the model effectively protects privacy while demonstrating good performance. Our experimental analysis demonstrates that on four standard datasets, our model achieves the same accuracy as classical neural network (NN) while utilizing only 22% of the parameters of NN, and exhibiting lower performance loss compared to other QNN.
This paper explores a novel approach to differential privacy utilizing quantum entanglement. Traditional differential privacy relies on adding noise to queries to obscure individual data contributions, a technique often limited by the dimensionality of the data and the potential for sophisticated attacks. We propose encoding data points into entangled qubit states, thereby leveraging the inherent correlations provided by entanglement to achieve stronger privacy guarantees. The core idea involves performing computations on these entangled states, minimizing the need for direct data exposure. Crucially, we introduce a mechanism for carefully calibrating measurement noise, exploiting the unique properties of entanglement to control the privacy loss. Our framework demonstrates the potential to significantly improve privacy levels compared to classical methods, especially when dealing with high-dimensional datasets, offering a new paradigm for privacy-preserving data analysis in the quantum realm. The theoretical analysis presents a novel framework for quantifying privacy loss, considering both the entanglement resource requirements and the calibrated noise levels. This work lays the foundation for future research into quantum-enhanced differential privacy and its implications for secure data processing.
The exponential growth of genomic data has created opportunities for collaborative biomedical research while posing significant privacy and security challenges. Traditional centralized machine learning approaches risk data breaches and regulatory non-compliance. This study proposes a novel hybrid architecture combining Federated Quantum Machine Learning (FQML) with Differential Privacy (DP) for secure multi-party genomic data analysis. Our simulated framework enables institutions to collaboratively train quantum machine learning models on distributed genomic datasets without exposing sensitive data. Using variational quantum circuits within a federated learning framework, we implement privacy-preserving models across three simulated genomic data silos. Gaussian noise injection via DP mechanisms during gradient updates ensures formal privacy guarantees (ε = 0.85). Experimental results on synthetic datasets demonstrate that FQML achieves 93.4% predictive accuracy in genome-wide association studies simulation, outperforming classical federated models in convergence speed by 27% and reducing communication overhead by 19%. These findings suggest FQML with DP provides a promising foundation for future secure genomic collaboration, though real-world validation remains necessary.
With the rapid advancement of information technology, data sharing has become increasingly accessible, leading to a heightened need for robust personal data protection. One important application in privacy-preserving computing is the aggregation of information when collaboratively establishing AI models through public distributed networks. To counter the threat posed by quantum computing to encrypted data, various quantum private summation (QPS) protocols have been proposed thus far. However, some of these existing protocols operate solely under modulo 2, while other approaches for modulo d often rely on impractically high-dimensional qudits. Therefore, this study proposes an innovative multiparty QPS method that balances participant data sharing and privacy without requiring high-dimensional photons. The proposed QPS protocol enables participants to contribute aggregated information to third parties without disclosing individual data. A security analysis further demonstrates that the proposed QPS effectively counters common eavesdropping attacks, ensuring reliable protection of personal data.