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

IEEE Transactions on Information Forensics and Security

Robust Quantum Federated Learning Against Colluding and Non-Colluding Byzantine Attacks

Jinjing Shi, Xuanli Lyu, Tian Chen, Shichao Zhang, Xuelong Li

Published Jan 1, 2026• 0 citations
Publisher
DOI
10.1109/tifs.2026.3714274

Related papers

Differentially-Private Quantum Federated Learning Against Gradient Inversion Attacks

Chaemoon Im, Soohyun Park, Joongheon Kim

IEEE Transactions on Networking•Jan 1, 2026•0 citations

A secure quantum federated learning for lightweight quantum clients

Wei‐Min Shi, Zi-Xuan Xu, Yu‐Guang Yang +1 more

Chinese Journal of Physics•Jul 6, 2026•0 citations

Defending Against Byzantine Attacks in Quantum Federated Learning

Qi Xia, Zeyi Tao, Qun Li

By combining the advantages of both quantum computing and deep learning, quantum neural networks have become popular in recent research. In order to collaborate multiple quantum machines with local training data to train a global model, quantum federated learning is proposed. However, similar to classic federated learning, when communicating with multiple machines, quantum federated learning also faces the threats of Byzantine attacks. The byzantine attack is a kind of attack in a distributed system when some machines upload malicious information instead of the honest computational results to the server. In this article, we compare the differences of Byzantine problems between classic distributed learning and quantum federated learning, and modify the previously proposed four kinds of Byzantine tolerant algorithms to the quantum version. We conduct simulated experiments to show a similar performance of the quantum version with the classic version.

International Conference on Mobility, Sensing and Networking (MSN)•Dec 1, 2021•25 citations

Quantum Federated Lasso With Anonymous Consensus Built Upon Quantum Blockchain

Liwei Lin, Yue Wu, Chuan Huang +3 more

ABSTRACT Objective With the rapid advancement of quantum computing technologies, Quantum Federated Learning (QFL) has emerged as a promising framework for privacy‐preserving quantum machine learning (QML). QFL provides a promising approach for privacy‐preserving quantum machine learning by enabling collaborative model training without sharing raw data. However, malicious or unreliable quantum clients may submit poisoned or low‐quality models, threatening the reliability of global aggregation. This study aims to develop a secure and robust QFL framework against such adversarial behaviors. Methods We propose a novel Quantum Federated Lasso (QFLasso) framework built on a quantum blockchain infrastructure. Local quantum models are trained with Lasso regularization to achieve sparse representations, and model updates are recorded on the blockchain. Quantum validators verify the quality and integrity of submitted models before aggregation. An anonymous quantum consensus mechanism is further introduced to ensure reliable agreement while preserving validator anonymity. Result The proposed framework integrates quantum Lasso optimization, federated learning, blockchain‐based verification, and anonymous quantum consensus to enhance the security, robustness, and privacy preservation of distributed quantum learning systems. The validator verification mechanism effectively filters unreliable model updates, while the anonymous consensus protocol improves resistance against dishonest or colluding validators and prevents targeted attacks on individual participants. Conclusion QFLasso provides a secure and scalable architecture for privacy‐preserving quantum intelligence, offering potential applications in secure quantum finance, distributed quantum sensing, and collaborative quantum medical diagnosis.

Software Practice and Experience•Jul 18, 2026•0 citations

Quantum Federated Learning: A Comprehensive Review and Analysis

Dev Gurung, Shiva Raj Pokhrel, Gang Li

IEEE Transactions on Artificial Intelligence•Jul 13, 2026•0 citations

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

Arnab Nath, Harsh Kasyap

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.

arXiv:2602.22269•Feb 25, 2026•0 citations
By combining the advantages of both quantum computing and deep learning, quantum neural networks have become popular in recent research. In order to collaborate multiple quantum machines with local training data to train a global model, quantum federated learning is proposed. However, similar to classic federated learning, when communicating with multiple machines, quantum federated learning also faces the threats of Byzantine attacks. The byzantine attack is a kind of attack in a distributed system when some machines upload malicious information instead of the honest computational results to the server. In this article, we compare the differences of Byzantine problems between classic distributed learning and quantum federated learning, and modify the previously proposed four kinds of Byzantine tolerant algorithms to the quantum version. We conduct simulated experiments to show a similar performance of the quantum version with the classic version.
ABSTRACT Objective With the rapid advancement of quantum computing technologies, Quantum Federated Learning (QFL) has emerged as a promising framework for privacy‐preserving quantum machine learning (QML). QFL provides a promising approach for privacy‐preserving quantum machine learning by enabling collaborative model training without sharing raw data. However, malicious or unreliable quantum clients may submit poisoned or low‐quality models, threatening the reliability of global aggregation. This study aims to develop a secure and robust QFL framework against such adversarial behaviors. Methods We propose a novel Quantum Federated Lasso (QFLasso) framework built on a quantum blockchain infrastructure. Local quantum models are trained with Lasso regularization to achieve sparse representations, and model updates are recorded on the blockchain. Quantum validators verify the quality and integrity of submitted models before aggregation. An anonymous quantum consensus mechanism is further introduced to ensure reliable agreement while preserving validator anonymity. Result The proposed framework integrates quantum Lasso optimization, federated learning, blockchain‐based verification, and anonymous quantum consensus to enhance the security, robustness, and privacy preservation of distributed quantum learning systems. The validator verification mechanism effectively filters unreliable model updates, while the anonymous consensus protocol improves resistance against dishonest or colluding validators and prevents targeted attacks on individual participants. Conclusion QFLasso provides a secure and scalable architecture for privacy‐preserving quantum intelligence, offering potential applications in secure quantum finance, distributed quantum sensing, and collaborative quantum medical diagnosis.
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.