TL;DR
The proposed approach displays improvement in accuracy and uniformity across client accuracy by enhancing the fairness of the federated learning system using a novel distribution-aware algorithm.
Local Inference Guided Aggregation for Heterogeneous Training Environments to Yield Enhancement Through Agreement and Regularization (LIGHTYEAR), a federated learning framework that performs update selection in function space using an NTK-based agreement score to characterize predictive behavior and determine a personalized aggregation set for each client.
Federated Learning (FL) enables collaborative model training without sharing raw data, but its effectiveness degrades under Non-IID client data and inefficient participation. In such settings, FedAvg may exhibit slow convergence, high variance across rounds, and elevated communication overhead. We propose a clustered ensemble framework to improve training stability and communication efficiency under label-skewed Non-IID distributions. Clients are clustered by data similarity, and one model is trained per cluster. At inference time, we employ a confidence-based ensemble that selects (or combines) cluster models, while adaptive client selection regulates participation within clusters to reduce redundant communication. Experiments on MNIST, Fashion-MNIST, CIFAR-10, and SVHN show more stable learning dynamics and lower communication cost while maintaining competitive accuracy compared to FedAvg. We also analyze how participation policies affect convergence behavior and client fairness.
Federated Learning (FL) enables collaborative model training across distributed clients without sharing raw data, but its performance degrades significantly under heterogeneous data distributions. While Clustered Federated Learning (CFL) mitigates heterogeneity by grouping similar clients, most existing methods primarily address label distribution skew and overlook feature distribution skew, where clients share labels but differ systematically in input characteristics. In this paper, we reveal a key observation: under feature distribution skew, different network layers exhibit highly unequal sensitivity, with early layers encoding strongly client-specific feature patterns that dominate aggregation behavior. Motivated by this, we propose SensCluster, a novel sensitivity-aware CFL framework that constructs compact client representations by selecting parameters that are most responsive to local feature distributions. By leveraging Fisher information, SensCluster identifies and retains only the most influential parameters, constructing sensitivity-aware representations for clustering while reducing interference from irrelevant features. Extensive experiments on both image (FEMNIST, CIFAR10-c, CINIC10-c, EuroSAT) and time-series (HAR) benchmarks demonstrate that SensCluster consistently outperforms state-of-the-art CFL methods across diverse feature skew scenarios.
The process of selecting suitable clients to participate in Federated Learning (FL) remains a critical challenge due to non-IID heterogeneity in data distributions and heterogeneity in computational resources among participating clients. Existing client selection approaches typically assess clients individually using attributes such as model accuracy. Such individual-based client assessment ignores group effects, which are paramount for global model performance. In this paper, we propose a Genetic-Algorithm (GA) based client selection mechanism that is applicable to both horizontal FL (HFL) and vertical FL (VFL). Candidate client groups are evaluated by a multi-criteria fitness function that jointly models group data size, feature coverage, label coverage, class balance, feature-distribution divergence, feature importance, computational power, reputation, accuracy, and outlier proportion. The GA performs an efficient search over the combinatorial space of client subsets. The proposed approach is evaluated using image classification, diabetes prediction, and rain prediction tasks. The proposed approach improves global model accuracy and accelerates convergence. On the MNIST dataset, it peaks at 98.37%, outperforming DSCS (97.27%) and FedMCCS (97.07%). On the diabetes dataset, it achieves 88.37%, while DSCS and FedMCCS achieve 87.15% and 86.64%, respectively. On the rain prediction task, it attains a peak accuracy of 83.57% and converges to 83.36%, compared with 81.98% for FedMCCS and 80.38% for DSCS.
Federated learning (FL) is a popular distributed learning framework where multiple clients perform local training and a server aggregates the locally updated models. FL enables decentralized training while preserving the privacy of clients'datasets. However, non-independent and identically distributed (non-IID) or noisy datasets can lead to low model accuracy or high convergence latency. Precluding these clients through client selection may mitigate the problem, but heavily biased client selections may also degrade the learning performance. In this study, we first experimentally measure the impact of non-IID data (including skews in data quantity and label distribution), noisy data, and fairness in client selection on model accuracy and convergence. We then propose a privacy-preserving scoring method to assess each client's contribution in FL, with experiments conducted to demonstrate the effectiveness of the proposed assessment.