arXiv:2609.00356

A Stable Aggregation Method for Quantum Federated Learning

Shanika Iroshi Nanayakkara, Shiva Raj Pokhrel

Published Jun 25, 2026• 0 citations

Abstract

Abstract—Quantum federated learning (QFL) enables clients to train quantum neural network (QNN) models without sharing private data. We find that aggregation in QFL is unstable under heterogeneous data, unreliable communication, variable fidelity, latency, and quantum hardware noise. Moreover, QFL is non-trivial to address because several QNN parameters are periodic angles, and Euclidean averaging often fails to capture the inherent dynamics. We develop a novel self-consistent midpoint aggregation method for the stable design and implementation of QFLs. We combine QoS-aware client weighting, circular parameter aggregation, and bounded midpoint-based update control. We perform several angular tests and experiments on IBM's real Quantum machines for validation, confirming our approach. Extensive evaluations and experiments on medical and financial datasets show improved stability, lower volatility, and competitive accuracy. Index Terms—Quantum federated learning, adaptive aggregation, geometry-aware aggregation, self-consistent midpoint, quality of service, angular aggregation, quantum neural networks.

DOI: 10.48550/arxiv.2609.00356

Related papers

A Drift Stable Quantum Federated Learning for Intelligent Services

Shanika Iroshi Nanayakkara, Shiva Raj Pokhrel

Quantum federated learning enables distributed clients to train quantum neural networks without sharing local data, making it promising for privacy-aware intelligent services. Intelligent services in this context refer to privacy-sensitive distributed decision systems, such as fraud detection and genomic classification, where reliable and fair client-level learning is as important as the accuracy of the aggregate model. However, heterogeneous client data and noisy quantum optimization often cause unstable local updates, client drift, and unfair performance between clients. This paper proposes DUQFL-Prox, a drift-stable quantum federated learning framework based on deep-unfolded local optimization. Instead of using a fixed local optimizer, each client performs adaptive unfolded SPSA updates, while a proximal term keeps the local model close to the global model. A lightweight controller learns step-specific optimization parameters to improve post-aggregation performance. Experiments on financial fraud and genomic classification tasks show that DUQFL-Prox improves stability, generalization, and client fairness compared with standard QFL baselines. The results suggest that deep-unfolded quantum federated learning can support more reliable and fair intelligent services in heterogeneous distributed environments.

arXiv:2607.21647Jul 22, 20260 citations

Toward Uniform Quantum Federated Aggregation: Heterogeneity Exclusion Using Entropy and Fidelity

Seok Bin Son, Soohyun Park

Quantum federated learning (QFL) incorporates the principles of quantum neural networks (QNNs) and federated learning (FL), ensuring privacy by updating quantum parameters to the server. However, server aggregation in general QFL involves quantum parameters from all local models, leading to performance degradation due to the participation of outlier heterogeneous local models. Inspired by this, a novel QFL algorithm is proposed to enhance learning performance by excluding outliers before aggregation using Lyapunov optimization at QFL server. This Lyapunov optimization is used to take care of the tradeoff between accuracy and latency, where the accuracy is associated with the freshness of aggregation. Additionally, an entropy-and-fidelity aware algorithm is proposed, which relies on the degree control to select outliers for exclusion comprehensively. This algorithm addresses data and quantum aspects by evaluating the imbalance of data classes in each local model using entropy and quantifying the dissimilarity in quantum states between the target model and each local model through fidelity. Experimental results demonstrate that the proposed algorithm outperforms benchmarks, effectively excluding heterogeneous local models to improve performance while ensuring stability.

IEEE Internet of Things JournalOct 30, 20248 citations

A2G-QFL: Adaptive Aggregation with Two Gains in Quantum Federated Learning

Shanika Iroshi Nanayakkara, Shiva Raj Pokhrel

Federated learning (FL) deployed over quantumenabled and heterogeneous classical networks can degrade under uneven client quality, modeled teleportation/link fidelity, and communication instability, while standard aggregation rules assume Euclidean topology and uniform reliability, limiting their suitability for emerging quantum federated systems. We introduce A2G (Adaptive Aggregation with Two Gains). This dualgain framework jointly regulates server-side relaxation through a geometry gain. It modulates client importance using a QoS gain derived from teleportation/link fidelity, latency, and instability (variance of the QoS signal). The proposed A2G update rule shows that A2G recovers baseline approaches such as FedAvg, Latency-only, QoS-aware averaging, and manifold-inspired aggregation as special cases. Experiments on a Qiskit Aer-based quantum-classical simulation testbed demonstrate improved stability and accuracy under heterogeneous, realistic, and simulated quantum-noisy conditions.

International Conference on Quantum Communications, Networking, and Computing (QCNC)Apr 6, 20260 citations

A Drift-Stable Quantum Federated Learning for Intelligent Services

Shanika Iroshi Nanayakkara, Shiva Raj Pokhrel

Quantum federated learning enables distributed clients to train quantum neural networks without sharing local data, making it promising for privacy-aware intelligent services. Intelligent services in this context refer to privacy-sensitive distributed decision systems, such as fraud detection and genomic classification, where reliable and fair client-level learning is as important as the accuracy of the aggregate model. However, heterogeneous client data and noisy quantum optimization oftencause unstable local updates, client drift, and unfair performance between clients. This paper proposes DUQFL-Prox, a drift-stable quantum federated learning framework based on deep-unfolded local optimization. Instead of using a fixed local optimizer, each client performs adaptive unfolded SPSA updates, while aproximal term keeps the local model close to the global model. A lightweight controller learns step-specific optimization parameters to improve post-aggregation performance. Experiments on financial fraud and genomic classification tasks show that DUQFL-Prox improves stability, generalization, and client fairnesscompared with standard QFL baselines. The results suggest that deep-unfolded quantum federated learning can support more reliable and fair intelligent services in heterogeneous distributed environments.

Jul 14, 20260 citations

Quantum Federated Aggregation using Joint Fidelity-and-Entropy Computation

Seok Bin Son, Soohyun Park

Quantum federated learning (QFL) incorporates the principles of quantum neural networks (QNN) over federated learning (FL). It updates quantum parameters to the server. However, QFL aggregation involves quantum parameters from all local QNN models, leading to performance degradation due to outlier models. Thus, a novel fair QFL algorithm is proposed to enhance learning performance by eliminating outliers using Lyapunov optimization at the QFL server. This approach is used to take care of the trade-off between accuracy and latency, where the accuracy is associated with the freshness of aggregation. Moreover, a fidelity/entropy-adaptive algorithm is proposed, which relies on the degree control to select outliers. This approach (i) quantifies the dissimilarity in quantum states between the target model and each local device using fidelity, and (ii) evaluates the imbalance of data classes in each local device using entropy. Experimental results demonstrate that the proposed fair QFL algorithm outperforms benchmarks.

IEEE International Conference on Distributed Computing Systems Workshops (ICDCSW)Jul 21, 20250 citations

DAQFL: Dynamic Aggregation Quantum Federated Learning Algorithm for Intelligent Diagnosis in Internet of Medical Things

Zhiguo Qu, Xuemeng Zhao, Le Sun +1 more

Federated learning (FL) is a privacy-preserving alternative to centralized machine learning, where model training is performed on local devices and only global model updates are shared, effectively addressing challenges, such as data silos and privacy protection. Recently, quantum FL (QFL), an emerging FL branch, has garnered significant attention in many industry applications. However, existing QFL algorithms predominantly employ average weighting for global model training, which shows poor performance on heterogeneous healthcare data. To address this challenge, this study proposes a dynamic aggregation QFL algorithm (DAQFL) for intelligent diagnosis. Specifically, it utilizes quantum neural networks (QNNs) as local training models and designs corresponding variational quantum circuits (VQC). To mitigate performance degradation caused by the heterogeneity of medical industrial data, a dynamic aggregation method based on accuracy is proposed to enhance global model performance effectively. Extensive experiments with three distribution settings, including independent and identically distributed (IID), non-independent and identically distributed (Non-IID), and long-tail datasets, show that DAQFL outperforms baseline algorithms in accuracy and training speed. It also performs well in privacy protection and robustness of anti-noise, improving its suitability for real-world medical applications.

IEEE Internet of Things JournalJan 31, 202523 citations