A novel self-consistent midpoint aggregation method is developed that combines QoS-aware client weighting, circular parameter aggregation, and bounded midpoint-based update control and shows improved stability, lower volatility, and competitive accuracy in QFL.
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-trivially challenging because several QNN parameters are periodic angles, where Euclidean averaging often fails to capture the inherent dynamics. We develop a novel self-consistent midpoint aggregation method for stable QFL design and implementation. We combine QoS-aware client weighting, circular parameter aggregation, and bounded midpoint-based update control. We perform several angular tests and IBM real Quantum machines experiments for validation confirming our approach. Extensive evaluations and experiments on medical and financial datasets show improved stability, lower volatility, and competitive accuracy.
KWA is proposed, a hyperparameter-free aggregation strategy that recovers FedAvg under IID conditions, adapts to client drift, and requires no manually tuned hyperparameters.
For quantum machine learning, the exact boundary between classical and quantum advantage is still poorly understood. Direct comparison between quantum neural networks (QNNs) and existing classical models, which encompass fundamentally different function classes, often fails to provide broader insight into the differenc...
Oliver Knitter, Jonathan Mei, Sang Hyub Kim et al.· 0 citations
The practical value of quantum machine learning for network anomaly detection remains unclear under controlled and comparable benchmarking conditions. This study evaluates a trash-qubit-based Quantum Autoencoder (QAE) against seven classical anomaly-detection methods on UNSW-NB15 using shared data partitions, preproces...
Gönül Altay, E. Küçüksille· Electronics· 0 citations
Electrocardiogram (ECG) recordings are sensitive biomedical data, limiting the ability of hospitals and wearable devices to share raw signals for centralized model training. Federated learning addresses this practical privacy constraint by enabling collaborative model training while keeping raw biosignal data at their...
Chun-Hua Lin, Samuel Yen-Chi Chen, Yu-Chao Hsu et al.· 0 citations
This work employs the Bures metric as a local preconditioner and uses the mean Uhlmann curvature to develop an achievable-precision aggregation rule that dynamically down-weights unreliable clients and establishes theoretical guarantees by proving a convergence theorem and a variance-dominance proposition.
Driven by the vision of intelligent and adaptive connectivity in the sixth-generation (6G) networks, future wireless systems are expected to support advanced applications such as cognitive resource allocation, semantic communication (SC), and low-latency integrated sensing and communication (ISAC) or digital twins. How...
Usama Inam Paracha, Uman Khalid, Syed Muhammad Abuzar Rizvi et al.· IEEE Transactions on Network...· 0 citations
Related blog posts
MIT News · Artificial Intelligence· news.mit.eduOct 13, 2026
Machine-learning networks can learn without detailed instructions from a designer. Those that do may offer insight into how the brain reorganizes itself during learning.
MIT News · Artificial Intelligence· news.mit.eduSep 29, 2026
Professor Sherry Turkle’s new book, “Artificial Intimacy,” offers a withering critique of chatbots and the antisocial dynamics she believes they encourage.
