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:2602.22269

CQSA: Byzantine-robust Clustered Quantum Secure Aggregation in Federated Learning

Arnab Nath, Harsh Kasyap

Published May 22, 2026• 2 citations
arXivHTMLPDF

Abstract

Federated Learning (FL) enables collaborative model training without sharing raw data. However, shared local model updates remain vulnerable to inference and poisoning attacks. Secure aggregation schemes have been proposed to mitigate these attacks. In this work, we aim to understand how these techniques are implemented in quantum-assisted FL. Quantum Secure Aggregation (QSA) has been proposed, offering information-theoretic privacy by encoding client updates into the global phase of multipartite entangled states. Existing QSA protocols, however, rely on a single global Greenberger–Horne–Zeilinger (GHZ) state shared among all participating clients. This design poses fundamental challenges: fidelity of large-scale GHZ states deteriorates rapidly with the increasing number of clients; and (ii) the global aggregation prevents the detection of Byzantine clients. We propose Clustered Quantum Secure Aggregation (CQSA), a modular aggregation framework that reconciles the physical constraints of near-term quantum hardware along with the need for Byzantine-robustness in FL. CQSA randomly partitions the clients into small clusters, each performing local quantum aggregation using high-fidelity, low-qubit GHZ states. The server analyzes statistical relationships between cluster-level aggregates employing common statistical measures such as cosine similarity and Euclidean distance to identify malicious contributions. Through theoretical analysis and simulations under depolarizing noise, we demonstrate that CQSA ensures stable model convergence, achieves superior state fidelity over global QSA.

DOI
10.1145/3803630.3809169

Related papers

Decentralized Federated Learning with Quantum Key Distribution for Secure Aggregation

Jincheng Zhang

Federated learning (FL) offers a promising approach to training machine learning models on decentralized data sources without direct data sharing, addressing privacy concerns. However, traditional FL is vulnerable to attacks targeting model updates, potentially compromising data privacy. This paper proposes a novel decentralized federated learning framework that integrates quantum key distribution (QKD) to establish secure keys for data aggregation. The core claim is that QKD significantly enhances the security and efficiency of FL. The proposed system utilizes QKD to generate and distribute shared keys between participating devices, enabling secure aggregation of model updates while mitigating the risk of individual data exposure. This approach directly addresses the security vulnerabilities inherent in standard FL. The framework's architecture relies on a decentralized network topology, further bolstering resilience against single points of failure. Mathematical formulations detailing key aspects of the system, including key distribution protocols, aggregation processes, and security analysis, are presented. This work represents a significant advancement in secure FL by combining the strengths of two cutting-edge technologies: FL and QKD.

•Sep 5, 2026•0 citations

Byzantine-Resilient Federated Learning via QUBO-Based Client Selection on Quantum Annealers

Andras Ferenczi, Sutapa Samanta, Dagen Wang +1 more

Federated Learning (FL) trains a global model across decentralized clients while preserving data privacy, but at scale it is vulnerable to malicious updates. Byzantine-resilient aggregation methods such as MultiKrum score gradients against their nearest neighbors and can miss malicious updates that preserve the statistical properties of honest ones. We propose a quantum annealing approach that reformulates client selection as a Quadratic Unconstrained Binary Optimization (QUBO) problem, encoding pairwise distances into a cost function solved by quantum annealers (QA). Unlike MultiKrum's greedy per-client scoring, the QUBO formulation jointly optimizes over all subsets to find the mutually closest group of $m$ clients. At small scale (15 clients), QUBO outperforms MultiKrum on the most challenging Byzantine attacks: e.g., Advanced LIE is detected with 95.11% accuracy versus 81.33% on MNIST and 97.78% versus 75.56% on CIFAR-10. QUBO fares poorly on simpler attacks where MultiKrum excels, so the two methods are complementary. QUBO quality also degrades as the number of clients grows. To address this, we introduce a MultiSignal ensemble that uses a dual-feature routing gate based on Euclidean and cosine Krum score gaps to classify attacks into four regimes and routes evasion attacks to a suspicion-penalized QUBO with agreement voting. At 100 clients on MNIST, MultiSignal achieves 95.3% average detection accuracy versus 91.8% for classical MultiKrum, with the largest gains on Sparse Lie (72.0% to 95.2%, +23.2 points) and Advanced Lie (80.4% to 85.2%, +4.8 points). These results show that QUBO-based quantum annealing with MultiSignal is a principled and scalable defense against the most challenging Byzantine strategies in federated learning.

arXiv:2605.16438•May 14, 2026•0 citations

Quantum Federated Learning: Bridging Quantum Computing and Distributed AI

Tariq Qayyum, Manzoor Ahmed Khan, Asadullah Tariq +4 more

In recent years, the growing demand for secure, efficient, and scalable machine learning has highlighted significant challenges in traditional Federated Learning (FL) systems. These challenges include computational inefficiencies in handling large-scale, high-dimensional data, privacy concerns, and the limitations of classical hardware in solving complex optimization problems. Additionally, the increasing complexity of data and the need for more robust privacy-preserving methods have pushed the boundaries of classical FL. To address these issues, this paper presents a detailed framework for Quantum Federated Learning (QFL), where quantum computing and FL converge to enable distributed quantum model training with enhanced privacy preservation. The system is modeled using local quantum models on clients and a global aggregation strategy at the server. We develop quantum gradient-based optimization with quantum cross-entropy loss, privacy preservation through encryption, and robustness to quantum noise. A set of theorems is provided to prove the correctness and efficiency of the model, supported by rigorous mathematical formulations. We used IBM Qiskit for the simulation and compared our proposed QFL with classical FL. The results demonstrate that our QFL outperformed classical FL in terms of accuracy, precision, recall, and F1 score.

IEEE/ACM International Conference on Utility and Cloud Computing (UCC)•Dec 16, 2024•6 citations

Beyond Classical Multi-Krum: Enhancing Byzantine Detection with Quantum Embeddings

Andras Ferenczi, Dagen Wang, Sutapa Samanta +1 more

Federated Learning (FL) is a promising method for privacy-preserving training of Machine Learning (ML) models across distributed devices (clients). Nevertheless, large-scale deployment of such processes on often untrusted devices can lead to the emergence of invalid updates that may compromise the global model. Numerous robust aggregation rules exist to mitigate such risks; however, opportunities remain for further improvement. The advent of quantum computers offers new avenues to revisit these aggregation methods and potentially deliver improved, more scalable solutions. In this paper, we introduce a novel technique based on computing quantum embeddings of client updates. The quantum embedding approach can extract richer information by projecting the data into a higher-dimensional space. We show empirically that this method outperforms MultiKrum in our experiments. This approach holds significant promise as quantum hardware evolves and the proposed hybrid solution scales.

IEEE International Conference on Quantum Artificial Intelligence (QAI)•Nov 2, 2025•1 citation

Federated Learning with Quantum Secure Aggregation

Yichi Zhang, Chao Zhang, Cai Zhang +3 more

This article illustrates a novel Quantum Secure Aggregation (QSA) scheme that is designed to provide highly secure and efficient aggregation of local model parameters for federated learning. The scheme is secure in protecting private model parameters from being disclosed to semi-honest attackers by utilizing quantum bits i.e. qubits to represent model parameters. The proposed security mechanism ensures that any attempts to eavesdrop private model parameters can be immediately detected and stopped. The scheme is also efficient in terms of the low computational complexity of transmitting and aggregating model parameters through entangled qubits. Benefits of the proposed QSA scheme are showcased in a horizontal federated learning setting in which both a centralized and decentralized architectures are taken into account. It was empirically demonstrated that the proposed QSA can be readily applied to aggregate different types of local models including logistic regression (LR), convolutional neural networks (CNN) as well as quantum neural network (QNN), indicating the versatility of the QSA scheme. Performances of global models are improved to various extents with respect to local models obtained by individual participants, while no private model parameters are disclosed to semi-honest adversaries.

arXiv:2207.07444•Jul 9, 2022•50 citations

Enhancing Federated Learning Privacy with QUBO

Andras Ferenczi, Sutapa Samanta, Dagen Wang +1 more

Federated learning (FL) is a widely used method for training machine learning (ML) models in a scalable way while preserving privacy (i.e., without centralizing raw data). Prior research shows that the risk of exposing sensitive data increases cumulatively as the number of iterations where a client's updates are included in the aggregated model increase. Attackers can launch membership inference attacks (MIA; deciding whether a sample or client participated), property inference attacks (PIA; inferring attributes of a client's data), and model inversion attacks (MI; reconstructing inputs), thereby inferring client-specific attributes and, in some cases, reconstructing inputs. In this paper, we mitigate risk by substantially reducing per client exposure using a quantum computing-inspired quadratic unconstrained binary optimization (QUBO) formulation that selects a small subset of client updates most relevant for each training round. In this work, we focus on two threat vectors: (i) information leakage by clients during training and (ii) adversaries who can query or obtain the global model. We assume a trusted central server and do not model server compromise. This method also assumes that the server has access to a validation/test set with global data distribution. Experiments on the MNIST dataset with 300 clients in 20 rounds showed a 95.2% per-round and 49% cumulative privacy exposure reduction, with 147 clients' updates never being used during training while maintaining in general the full-aggregation accuracy or even better. The method proved to be efficient at lower scale and more complex model as well. A CINIC-10 dataset-based experiment with 30 clients resulted in 82% per-round privacy improvement and 33% cumulative privacy.

cs.LGCross-listedarXiv:2511.02785•Nov 5, 2025
Federated Learning (FL) enables collaborative model training without sharing raw data. However, shared local model updates remain vulnerable to inference and poisoning attacks. Secure aggregation schemes have been proposed to mitigate these attacks. In this work, we aim to understand how these techniques are implemented in quantum-assisted FL. Quantum Secure Aggregation (QSA) has been proposed, offering information-theoretic privacy by encoding client updates into the global phase of multipartite entangled states. Existing QSA protocols, however, rely on a single global Greenberger–Horne–Zeilinger (GHZ) state shared among all participating clients. This design poses fundamental challenges: fidelity of large-scale GHZ states deteriorates rapidly with the increasing number of clients; and (ii) the global aggregation prevents the detection of Byzantine clients. We propose Clustered Quantum Secure Aggregation (CQSA), a modular aggregation framework that reconciles the physical constraints of near-term quantum hardware along with the need for Byzantine-robustness in FL. CQSA randomly partitions the clients into small clusters, each performing local quantum aggregation using high-fidelity, low-qubit GHZ states. The server analyzes statistical relationships between cluster-level aggregates employing common statistical measures such as cosine similarity and Euclidean distance to identify malicious contributions. Through theoretical analysis and simulations under depolarizing noise, we demonstrate that CQSA ensures stable model convergence, achieves superior state fidelity over global QSA.
Federated learning (FL) offers a promising approach to training machine learning models on decentralized data sources without direct data sharing, addressing privacy concerns. However, traditional FL is vulnerable to attacks targeting model updates, potentially compromising data privacy. This paper proposes a novel decentralized federated learning framework that integrates quantum key distribution (QKD) to establish secure keys for data aggregation. The core claim is that QKD significantly enhances the security and efficiency of FL. The proposed system utilizes QKD to generate and distribute shared keys between participating devices, enabling secure aggregation of model updates while mitigating the risk of individual data exposure. This approach directly addresses the security vulnerabilities inherent in standard FL. The framework's architecture relies on a decentralized network topology, further bolstering resilience against single points of failure. Mathematical formulations detailing key aspects of the system, including key distribution protocols, aggregation processes, and security analysis, are presented. This work represents a significant advancement in secure FL by combining the strengths of two cutting-edge technologies: FL and QKD.
Federated Learning (FL) trains a global model across decentralized clients while preserving data privacy, but at scale it is vulnerable to malicious updates. Byzantine-resilient aggregation methods such as MultiKrum score gradients against their nearest neighbors and can miss malicious updates that preserve the statistical properties of honest ones. We propose a quantum annealing approach that reformulates client selection as a Quadratic Unconstrained Binary Optimization (QUBO) problem, encoding pairwise distances into a cost function solved by quantum annealers (QA). Unlike MultiKrum's greedy per-client scoring, the QUBO formulation jointly optimizes over all subsets to find the mutually closest group of mmm clients. At small scale (15 clients), QUBO outperforms MultiKrum on the most challenging Byzantine attacks: e.g., Advanced LIE is detected with 95.11% accuracy versus 81.33% on MNIST and 97.78% versus 75.56% on CIFAR-10. QUBO fares poorly on simpler attacks where MultiKrum excels, so the two methods are complementary. QUBO quality also degrades as the number of clients grows. To address this, we introduce a MultiSignal ensemble that uses a dual-feature routing gate based on Euclidean and cosine Krum score gaps to classify attacks into four regimes and routes evasion attacks to a suspicion-penalized QUBO with agreement voting. At 100 clients on MNIST, MultiSignal achieves 95.3% average detection accuracy versus 91.8% for classical MultiKrum, with the largest gains on Sparse Lie (72.0% to 95.2%, +23.2 points) and Advanced Lie (80.4% to 85.2%, +4.8 points). These results show that QUBO-based quantum annealing with MultiSignal is a principled and scalable defense against the most challenging Byzantine strategies in federated learning.
In recent years, the growing demand for secure, efficient, and scalable machine learning has highlighted significant challenges in traditional Federated Learning (FL) systems. These challenges include computational inefficiencies in handling large-scale, high-dimensional data, privacy concerns, and the limitations of classical hardware in solving complex optimization problems. Additionally, the increasing complexity of data and the need for more robust privacy-preserving methods have pushed the boundaries of classical FL. To address these issues, this paper presents a detailed framework for Quantum Federated Learning (QFL), where quantum computing and FL converge to enable distributed quantum model training with enhanced privacy preservation. The system is modeled using local quantum models on clients and a global aggregation strategy at the server. We develop quantum gradient-based optimization with quantum cross-entropy loss, privacy preservation through encryption, and robustness to quantum noise. A set of theorems is provided to prove the correctness and efficiency of the model, supported by rigorous mathematical formulations. We used IBM Qiskit for the simulation and compared our proposed QFL with classical FL. The results demonstrate that our QFL outperformed classical FL in terms of accuracy, precision, recall, and F1 score.
Federated Learning (FL) is a promising method for privacy-preserving training of Machine Learning (ML) models across distributed devices (clients). Nevertheless, large-scale deployment of such processes on often untrusted devices can lead to the emergence of invalid updates that may compromise the global model. Numerous robust aggregation rules exist to mitigate such risks; however, opportunities remain for further improvement. The advent of quantum computers offers new avenues to revisit these aggregation methods and potentially deliver improved, more scalable solutions. In this paper, we introduce a novel technique based on computing quantum embeddings of client updates. The quantum embedding approach can extract richer information by projecting the data into a higher-dimensional space. We show empirically that this method outperforms MultiKrum in our experiments. This approach holds significant promise as quantum hardware evolves and the proposed hybrid solution scales.
This article illustrates a novel Quantum Secure Aggregation (QSA) scheme that is designed to provide highly secure and efficient aggregation of local model parameters for federated learning. The scheme is secure in protecting private model parameters from being disclosed to semi-honest attackers by utilizing quantum bits i.e. qubits to represent model parameters. The proposed security mechanism ensures that any attempts to eavesdrop private model parameters can be immediately detected and stopped. The scheme is also efficient in terms of the low computational complexity of transmitting and aggregating model parameters through entangled qubits. Benefits of the proposed QSA scheme are showcased in a horizontal federated learning setting in which both a centralized and decentralized architectures are taken into account. It was empirically demonstrated that the proposed QSA can be readily applied to aggregate different types of local models including logistic regression (LR), convolutional neural networks (CNN) as well as quantum neural network (QNN), indicating the versatility of the QSA scheme. Performances of global models are improved to various extents with respect to local models obtained by individual participants, while no private model parameters are disclosed to semi-honest adversaries.
Federated learning (FL) is a widely used method for training machine learning (ML) models in a scalable way while preserving privacy (i.e., without centralizing raw data). Prior research shows that the risk of exposing sensitive data increases cumulatively as the number of iterations where a client's updates are included in the aggregated model increase. Attackers can launch membership inference attacks (MIA; deciding whether a sample or client participated), property inference attacks (PIA; inferring attributes of a client's data), and model inversion attacks (MI; reconstructing inputs), thereby inferring client-specific attributes and, in some cases, reconstructing inputs. In this paper, we mitigate risk by substantially reducing per client exposure using a quantum computing-inspired quadratic unconstrained binary optimization (QUBO) formulation that selects a small subset of client updates most relevant for each training round. In this work, we focus on two threat vectors: (i) information leakage by clients during training and (ii) adversaries who can query or obtain the global model. We assume a trusted central server and do not model server compromise. This method also assumes that the server has access to a validation/test set with global data distribution. Experiments on the MNIST dataset with 300 clients in 20 rounds showed a 95.2% per-round and 49% cumulative privacy exposure reduction, with 147 clients' updates never being used during training while maintaining in general the full-aggregation accuracy or even better. The method proved to be efficient at lower scale and more complex model as well. A CINIC-10 dataset-based experiment with 30 clients resulted in 82% per-round privacy improvement and 33% cumulative privacy.