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:2507.07316 · IEEE International Conference on Computer Vision Workshops

AdeptHEQ-FL: Adaptive Homomorphic Encryption for Federated Learning of Hybrid Classical-Quantum Models with Dynamic Layer Sparing

Md Abrar Jahin, Taufikur Rahman Fuad, M. F. Mridha, Nafiz Fahad, Md. Jakir Hossen

Published Oct 19, 2025• 1 citation
arXivHTMLPDF

Abstract

Federated Learning (FL) faces inherent challenges in balancing model performance, privacy preservation, and commu-nication efficiency, especially in non-IID decentralized environments. Recent approaches either sacrifice formal privacy guarantees, incur high overheads, or overlook quantum-enhanced expressivity. We introduce AdeptHEQ-FL, a unified hybrid classical-quantum FLframework that integrates (i) a hybrid CNN-PQC architecture for expressive decen-tralized learning, (ii) an adaptive accuracy-weighted aggre-gation scheme leveraging differentially private validation accuracies, (iii) selective homomorphic encryption (HE) for secure aggregation of sensitive model layers, and (iv) dynamic layer-wise adaptive freezing to minimize communication overhead while preserving quantum adaptability. We establish formal privacy guarantees, provide convergence analysis, and conduct extensive experiments on the CIFAR-10, SVHN, and Fashion-MNIST datasets. AdeptHEQ-FL achieves a ≈ 25.43% and ≈ 14.17% accuracy improvement over Standard-FedQNN and FHE-FedQNN, respectively, on the CIFAR-10 dataset. Additionally, it reduces communication overhead by freezing less important layers, demonstrating the efficiency and practicality of our privacy-preserving, resource-aware design for FL. Our code is publicly available at: https://github.com/Abrar2652/QML-FL.

DOI
10.1109/iccvw69036.2025.00064

Related papers

MQFL-FHE: Multimodal Quantum Federated Learning Framework with Fully Homomorphic Encryption

Siddhant Dutta, Nouhaila Innan, Sadok Ben Yahia +2 more

The integration of fully homomorphic encryption (FHE) in federated learning (FL) has led to significant advances in data privacy. However, during the aggregation phase, it often results in performance degradation of the aggregated model, hindering the development of robust representational generalization. In this work, we propose a novel multimodal quantum federated learning framework that utilizes quantum computing to counteract the performance drop resulting from FHE. For the first time in FL, our framework combines a multimodal quantum mixture of experts (MQMoE) model with FHE, incorporating multimodal datasets for enriched representation and task-specific learning. Our MQMoE framework enhances performance on multimodal datasets and combined genomics and brain MRI scans, especially for underrepresented categories. Our results also demonstrate that the quantum-enhanced approach mitigates the performance degradation associated with FHE and improves classification accuracy across diverse datasets, validating the potential of quantum interventions in enhancing privacy in FL.

arXiv:2412.01858•International Joint Conference on Neural Networks (IJCNN)•Jun 30, 2025•3 citations

Quantum-Enhanced Federated Learning for Realtime Consumer Sentiment Analysis: a Privacypreserving Hybrid Framework

Hongzhen Yu, Hazirah Bee Binti Yusof Ali, M. Kazem Chamran

As real-time data processing becomes central to industries like retail, e-commerce, and entertainment, ensuring both computational efficiency and data privacy is increasingly vital. This study proposes a novel hybrid framework that integrates quantum-enhanced transformers within a Federated Learning (FL) system to perform real-time consumer sentiment analysis. While FL ensures data privacy by decentralizing model training across edge devices, it struggles with issues like high communication overhead and slow convergence in non-IID environments. On the other hand, quantum-enhanced models offer computational acceleration but often lack sufficient privacy safeguards, especially during quantum circuit operations. This research addresses these challenges by developing a quantumenhanced transformer optimized for sentiment analysis and embedding it into a privacy-preserving FL architecture. Advanced encryption techniques such as differential privacy and quantum homomorphic encryption are incorporated to mitigate data leakage risks. The study adopts a four-phase methodology: model development using quantum simulation platforms (e.g., IBM Qiskit), FL integration using FedAvg protocols, system validation through benchmarking against classical models, and real-time deployment simulations across industry-specific datasets. Results show that the proposed system achieves superior performance in latency, model accuracy, and privacy metrics compared to traditional FL models. Despite challenges related to quantum noise and heterogeneity in client data, the system demonstrates robustness and scalability under real-time conditions. This research contributes to the field by providing a viable, secure, and scalable solution for deploying quantum-AI in decentralized environments. It opens pathways for future research in hybrid quantum-classical architectures and domain-specific AI applications that require enhanced privacy and speed.

International Conference on Artificial Intelligence, Robotics, and Communication (ICAIRC)•Nov 7, 2025•0 citations

Understanding the Resource Cost of Fully Homomorphic Encryption in Quantum Federated Learning

Lukas Böhm, Arjhun Swaminathan, A. Hannemann +1 more

Quantum Federated Learning (QFL) enables distributed training of Quantum Machine Learning (QML) models by sharing model gradients instead of raw data. However, these gradients can still expose sensitive user information. To enhance privacy, homomorphic encryption of parameters has been proposed as a solution in QFL and related frameworks. In this work, we evaluate the overhead introduced by Fully Homomorphic Encryption (FHE) in QFL setups and assess its feasibility for real-world applications. We implemented various QML models including a Quantum Convolutional Neural Network (QCNN) trained in a federated environment with parameters encrypted using the CKKS scheme. This work marks the first QCNN trained in a federated setting with CKKS-encrypted parameters. Models of varying architectures were trained to predict brain tumors from MRI scans. The experiments reveal that memory and communication overhead remain substantial, making FHE challenging to deploy. Minimizing overhead requires reducing the number of model parameters, which, however, leads to a decline in classification performance, introducing a trade-off between privacy and model complexity.

arXiv:2603.02799•Mar 3, 2026•0 citations

Quantum-enabled Federated Learning: Quantum Strategies for Performance Enhancement

Dongfen Li, Qiuyu Xiang, Peng Xiao +3 more

Federated Learning (FL) is a cutting-edge approach for distributed machine learning that demonstrates significant potential in addressing challenges such as data privacy protection and system heterogeneity. However, classical FL models often encounter performance bottlenecks when dealing with not independently identically distributed (non-IID) data distributions. Quantum machine learning (QML), with its unique advantages in improving computational efficiency and model representation, is a promising solution. This study systematically compares three FL architectures: a FL model based on classical CNNs (FedCNN), a quantum-classical hybrid FL model (FedQCNN) that processes input features using a quantum convolutional layer, and a FL model (FedCQNN) employing a quantum convolutional layer as an intermediate convolutional layer. These findings underscore the feasibility and efficacy of integrating QML into FL frameworks, opening the way for exploring innovative solutions in the field of distributed machine learning.

International Conference on Cloud Computing, Big Data Application and Software Engineering (CBASE)•Oct 24, 2025•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

CryptoQFL: Quantum Federated Learning on Encrypted Data

Cheng Chu, Lei Jiang, Fan Chen

Recent advancements in Quantum Neural Networks (QNNs) have demonstrated theoretical and experimental performance superior to their classical counterparts in a wide range of applications. However, existing centralized QNNs cannot solve many real-world problems because collecting large amounts of training data to a common public site is time-consuming and, more importantly, violates data privacy. Federated Learning (FL) is an emerging distributed machine learning framework that allows collaborative model training on decentralized data residing on multiple devices without breaching data privacy. Some initial attempts at Quantum Federated Learning (QFL) either only focus on improving the QFL performance or rely on a trusted quantum server that fails to preserve data privacy. In this work, we propose CryptoQFL, a QFL framework that allows distributed QNN training on encrypted data. CryptoQFL is (1) secure, because it allows each edge to train a QNN with local private data, and encrypt its updates using quantum homomorphic encryption before sending them to the central quantum server; (2) communication-efficient, as CryptoQFL quantize local gradient updates to ternary values, and only communicate non-zero values to the server for aggregation; and (3) computation-efficient, as CryptoQFL presents an efficient quantum aggregation circuit with significantly reduced latency compared to state-of-the-art approaches.

arXiv:2307.07012•IEEE International Conference on Quantum Computing and Engineering (QCE)•Sep 17, 2023•22 citations
Federated Learning (FL) faces inherent challenges in balancing model performance, privacy preservation, and commu-nication efficiency, especially in non-IID decentralized environments. Recent approaches either sacrifice formal privacy guarantees, incur high overheads, or overlook quantum-enhanced expressivity. We introduce AdeptHEQ-FL, a unified hybrid classical-quantum FLframework that integrates (i) a hybrid CNN-PQC architecture for expressive decen-tralized learning, (ii) an adaptive accuracy-weighted aggre-gation scheme leveraging differentially private validation accuracies, (iii) selective homomorphic encryption (HE) for secure aggregation of sensitive model layers, and (iv) dynamic layer-wise adaptive freezing to minimize communication overhead while preserving quantum adaptability. We establish formal privacy guarantees, provide convergence analysis, and conduct extensive experiments on the CIFAR-10, SVHN, and Fashion-MNIST datasets. AdeptHEQ-FL achieves a ≈ 25.43% and ≈ 14.17% accuracy improvement over Standard-FedQNN and FHE-FedQNN, respectively, on the CIFAR-10 dataset. Additionally, it reduces communication overhead by freezing less important layers, demonstrating the efficiency and practicality of our privacy-preserving, resource-aware design for FL. Our code is publicly available at: https://github.com/Abrar2652/QML-FL.
The integration of fully homomorphic encryption (FHE) in federated learning (FL) has led to significant advances in data privacy. However, during the aggregation phase, it often results in performance degradation of the aggregated model, hindering the development of robust representational generalization. In this work, we propose a novel multimodal quantum federated learning framework that utilizes quantum computing to counteract the performance drop resulting from FHE. For the first time in FL, our framework combines a multimodal quantum mixture of experts (MQMoE) model with FHE, incorporating multimodal datasets for enriched representation and task-specific learning. Our MQMoE framework enhances performance on multimodal datasets and combined genomics and brain MRI scans, especially for underrepresented categories. Our results also demonstrate that the quantum-enhanced approach mitigates the performance degradation associated with FHE and improves classification accuracy across diverse datasets, validating the potential of quantum interventions in enhancing privacy in FL.
As real-time data processing becomes central to industries like retail, e-commerce, and entertainment, ensuring both computational efficiency and data privacy is increasingly vital. This study proposes a novel hybrid framework that integrates quantum-enhanced transformers within a Federated Learning (FL) system to perform real-time consumer sentiment analysis. While FL ensures data privacy by decentralizing model training across edge devices, it struggles with issues like high communication overhead and slow convergence in non-IID environments. On the other hand, quantum-enhanced models offer computational acceleration but often lack sufficient privacy safeguards, especially during quantum circuit operations. This research addresses these challenges by developing a quantumenhanced transformer optimized for sentiment analysis and embedding it into a privacy-preserving FL architecture. Advanced encryption techniques such as differential privacy and quantum homomorphic encryption are incorporated to mitigate data leakage risks. The study adopts a four-phase methodology: model development using quantum simulation platforms (e.g., IBM Qiskit), FL integration using FedAvg protocols, system validation through benchmarking against classical models, and real-time deployment simulations across industry-specific datasets. Results show that the proposed system achieves superior performance in latency, model accuracy, and privacy metrics compared to traditional FL models. Despite challenges related to quantum noise and heterogeneity in client data, the system demonstrates robustness and scalability under real-time conditions. This research contributes to the field by providing a viable, secure, and scalable solution for deploying quantum-AI in decentralized environments. It opens pathways for future research in hybrid quantum-classical architectures and domain-specific AI applications that require enhanced privacy and speed.
Quantum Federated Learning (QFL) enables distributed training of Quantum Machine Learning (QML) models by sharing model gradients instead of raw data. However, these gradients can still expose sensitive user information. To enhance privacy, homomorphic encryption of parameters has been proposed as a solution in QFL and related frameworks. In this work, we evaluate the overhead introduced by Fully Homomorphic Encryption (FHE) in QFL setups and assess its feasibility for real-world applications. We implemented various QML models including a Quantum Convolutional Neural Network (QCNN) trained in a federated environment with parameters encrypted using the CKKS scheme. This work marks the first QCNN trained in a federated setting with CKKS-encrypted parameters. Models of varying architectures were trained to predict brain tumors from MRI scans. The experiments reveal that memory and communication overhead remain substantial, making FHE challenging to deploy. Minimizing overhead requires reducing the number of model parameters, which, however, leads to a decline in classification performance, introducing a trade-off between privacy and model complexity.
Federated Learning (FL) is a cutting-edge approach for distributed machine learning that demonstrates significant potential in addressing challenges such as data privacy protection and system heterogeneity. However, classical FL models often encounter performance bottlenecks when dealing with not independently identically distributed (non-IID) data distributions. Quantum machine learning (QML), with its unique advantages in improving computational efficiency and model representation, is a promising solution. This study systematically compares three FL architectures: a FL model based on classical CNNs (FedCNN), a quantum-classical hybrid FL model (FedQCNN) that processes input features using a quantum convolutional layer, and a FL model (FedCQNN) employing a quantum convolutional layer as an intermediate convolutional layer. These findings underscore the feasibility and efficacy of integrating QML into FL frameworks, opening the way for exploring innovative solutions in the field of distributed machine learning.
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.
Recent advancements in Quantum Neural Networks (QNNs) have demonstrated theoretical and experimental performance superior to their classical counterparts in a wide range of applications. However, existing centralized QNNs cannot solve many real-world problems because collecting large amounts of training data to a common public site is time-consuming and, more importantly, violates data privacy. Federated Learning (FL) is an emerging distributed machine learning framework that allows collaborative model training on decentralized data residing on multiple devices without breaching data privacy. Some initial attempts at Quantum Federated Learning (QFL) either only focus on improving the QFL performance or rely on a trusted quantum server that fails to preserve data privacy. In this work, we propose CryptoQFL, a QFL framework that allows distributed QNN training on encrypted data. CryptoQFL is (1) secure, because it allows each edge to train a QNN with local private data, and encrypt its updates using quantum homomorphic encryption before sending them to the central quantum server; (2) communication-efficient, as CryptoQFL quantize local gradient updates to ternary values, and only communicate non-zero values to the server for aggregation; and (3) computation-efficient, as CryptoQFL presents an efficient quantum aggregation circuit with significantly reduced latency compared to state-of-the-art approaches.