FedLECC: Cluster- and Loss-Guided Client Selection for Federated Learning under Non-IID Data

Daniel M. Jimenez-Gutierrez; Giovanni Giunta; Mehrdad Hassanzadeh; Aris Anagnostopoulos; Ioannis Chatzigiannakis; Andrea Vitaletti

cs.DCMar 9, 2026

FedLECC: Cluster- and Loss-Guided Client Selection for Federated Learning under Non-IID Data

Authors: Daniel M. Jimenez-Gutierrez, Giovanni Giunta, Mehrdad Hassanzadeh, Aris Anagnostopoulos, Ioannis Chatzigiannakis, Andrea Vitaletti

Organizations: Sapienza University of Rome Via Ariosto 25 Rome, 00185, Italy

Abstract

Federated Learning (FL) enables distributed Artificial Intelligence (AI) across cloud-edge environments by allowing collaborative model training without centralizing data. In cross-device deployments, FL systems face strict communication and participation constraints, as well as strong non-independent and identically distributed (non-IID) data that degrades convergence and model quality. Since only a subset of devices (a.k.a clients) can participate per training round, intelligent client selection becomes a key systems challenge. This paper proposes FedLECC (Federated Learning with Enhanced Cluster Choice), a lightweight, cluster-aware, and loss-guided client selection strategy for cross-device FL. FedLECC groups clients by label-distribution similarity and prioritizes clusters and clients with higher local loss, enabling the selection of a small yet informative and diverse set of clients. Experimental results under severe label skew show that FedLECC improves test accuracy by up to 12%, while reducing communication rounds by approximately 22% and overall communication overhead by up to 50% compared to strong baselines. These results demonstrate that informed client selection improves the efficiency and scalability of FL workloads in cloud-edge systems.

Figures & tables

Fig. 1: High-level client selection process in FedLECC.

Fig. 2: FL architecture with FedLECC client selection.

Dataset	#train	#test	#features	#classes
MNIST	60,000	10,000	784	10
FMNIST	60,000	10,000	784	10

TABLE I: Characteristics of the datasets used in our evaluation.

Fig. 3: Accuracy comparison for the baselines and FedLECC using FMNIST and

K=100

(best tuned hyperparameters).

	MNIST		FMNIST
Method	$K=100$	$K=250$	$K=100$	$K=300$
HD	0.90	0.86	0.90	0.86
Silhouette	0.641	0.502	0.723	0.409
FedAvg	0.681 $\pm$ 0.02	0.745 $\pm$ 0.03	0.565 $\pm$ 0.03	0.634 $\pm$ 0.03
FedProx	0.681 $\pm$ 0.03	0.745 $\pm$ 0.03	0.608 $\pm$ 0.02	0.636 $\pm$ 0.02
FedNova	0.696 $\pm$ 0.03	0.732 $\pm$ 0.03	0.554 $\pm$ 0.03	0.598 $\pm$ 0.04

TABLE II: Accuracy (mean

\pm

std) on MNIST and FMNIST under high non-IID data.

	MNIST		FMNIST
Method	$K=100$	$K=250$	$K=100$	$K=300$
FedAvg	49.28	123.21	49.28	126.28
FedProx	21.12	49.28	44.73	119.73
FedNova	17.70	132.63	73.61	93.92
FedDyn	41.31	38.05	15.22	35.61
HACCS	3.42	3.23	3.42	55.29

TABLE III: Average communication overhead (MB) on MNIST and FMNIST (smaller is better).

Client Selection in Federated Learning

Client selection in federated learning chooses which devices train each round to balance model quality, efficiency, fairness, and privacy.

See all 35 papers

Client and Training Data Selection for Computationally Efficient Synchronized Federated Learning
Sep 30, 2026Muzaffer Citir, Hiroki Nishikawa, Sangyoung ParkClient Selection in Federated Learning Non-IID Federated Learning
Smart Mobility Systems, Technical University of Berlin, Germany · Graduate School of Information Science and Technology, The University of Osaka, Japan
Assessing the Impacts of Imperfect Datasets on Client Selections in Federated Learning
Aug 3, 2026Yuan-Heng Tsai, Li-Hsing Yen, Yan-Wei ChenAlgorithmic Fairness Client Selection in Federated Learning
Department of Computer Science, National Yang Ming Chiao Tung University, Hsinchu, Taiwan.
Beyond Non-IID: Learner--Client Distribution Mismatch in Federated Learning
Aug 27, 2026Yiming Xie, Lili Su, Ningfang MiClient Selection in Federated Learning Non-IID Federated Learning
Department of Electrical and Computer Engineering Northeastern University, Boston, MA, USA

Sep 30, 2026J/K move · Enter open · S save

Client and Training Data Selection for Computationally Efficient Synchronized Federated Learning

Muzaffer Citir, Hiroki Nishikawa, Sangyoung Park

Smart Mobility Systems, Technical University of Berlin, Germany · Graduate School of Information Science and Technology, The University of Osaka, Japan

Client Selection in Federated Learning Non-IID Federated Learning Training Data Selection Federated Learning Heterogeneous Federated Learning

Open paper AskPDF arXiv

Federated learning (FL) is a promising paradigm of machine learning, which preserves user privacy by enabling learning without sharing raw data with a cloud server. Straggling clients have been a problem for FL as they introduce delays in aggregating the local models and hence, the convergence of the global model. Therefore, it is important to have a mechanism that ensures fast convergence of the global model as well as good FL participation rate. Another issue for the convergence of a model in FL is the non-independent and identically distributed (non-iid) data across the clients. Prior approaches based on probabilistic client selection do not work well under non-iid data especially when the number of clients is small. We show scenarios where such approaches fail and propose a joint client-training data selection algorithm for fast convergence of FL models. Our experiments on CIFAR-100 dataset show that convergence of the FL model can be significantly improved over prior works that can consider non-iid data and heterogeneous computation and higher model accuracy.

Abstract

Client Selection in Federated Learning

Explore similar work

Client and Training Data Selection for Computationally Efficient Synchronized Federated Learning

Assessing the Impacts of Imperfect Datasets on Client Selections in Federated Learning

Beyond Non-IID: Learner--Client Distribution Mismatch in Federated Learning