This special session will consider and discuss contributions with a particular focus on Federated Learning, a framework for decentralized, privacy-preserving machine learning. This session will address the discussion on Federated Learning approaches that extend, further enquire, or improve the existing Fuzzy Federated Learning theory. We will discuss the uses of Federated Learning models for new tasks in data science, artificial intelligence, deep learning, explainable/interpretable artificial intelligence, machine learning, decision-making, robotics, control, and information science. We will also consider contributions to novel applications of Federated Fuzzy models in real-world scenarios and research questions.
Topics of Interest
We invite submissions on all aspects of Fuzzy Federated Learning, including but not limited to:
• Theoretical Advances:
◦ Novel fuzzy aggregation operators for model fusion .
◦ Handling statistical challenges (non-IID data, concept drift) with fuzzy logic.
◦ Theoretical convergence analysis of fuzzy federated algorithms.
◦ Fusion of uncertain and imprecise information in FL settings.
◦ Fuzzy-based approaches for fairness and bias mitigation in FL.
• Privacy and Security:
◦ Integrating fuzzy logic with differential privacy and secure multi-party computation.
◦ Fuzzy-based methods for anomaly and intrusion detection in FL systems.
◦ Information fusion under privacy constraints.
• Novel Applications and Models:
◦ Federated Fuzzy Rule-Based Systems and Fuzzy Cognitive Maps.
◦ Federated Fuzzy Neural Networks.
◦ Applications in IoT, edge computing, robotics, control systems, and smart cities.
◦ Lightweight fuzzy models for resource-constrained edge devices.
◦ Management of heterogeneous edge device capabilities using fuzzy systems.
◦ Federated learning for interpretable and explainable AI (XAI) using fuzzy systems.
◦ Case studies in healthcare, industry 4.0, finance, and other domains.