Publications

Research papers, publications, and academic contributions

Peer-reviewed & talks

Journal articles

SPATL-XLC: Explainability-Driven Federated Learning Framework for Robust and Efficient Learning Under Non-IID Data

IEEE

Authors: Samuel Hailemariam Seifu, et al.

Venue: IEEE Access, 2025

DOI: 10.1109/ACCESS.2025.3589535

Abstract: This paper proposes an explainability-driven framework for robust and efficient federated learning under non-IID data distributions. The framework focuses on improving robustness, efficiency, and interpretability of federated learning systems.

Federated LearningExplainabilityNon-IID DataMachine Learning

View paper →