CfP Special Session @IJCNN2025: Towards Robust Federated Learning: Addressing Data and Device Heterogeneity

Dear Colleagues,
We invite you to submit your research to the Special Session “Towards Robust Federated Learning: Addressing Data and Device Heterogeneity” @ International Joint Conference on Neural Networks 2025, in Rome, June-July 2025 (date still to be defined).

Federated Learning (FL) has emerged as a transformative approach to enabling AI across decentralized and sensitive data sources, addressing privacy concerns in line with regulations like GDPR. However, practical deployment faces significant challenges due to non-iid (independent and identically distributed) data and device heterogeneity. These challenges impact FL’s scalability, fairness, and reliability, particularly in critical domains such as healthcare, IoT, and finance.

We welcome original research on (but not limited to):
* Algorithmic and Theoretical Advances: Novel algorithms improving FL’s performance and adaptability to non-iid data and heterogeneous devices.
* Fairness and Robustness: Strategies for creating equitable and resilient models across diverse data sources and device capacities.
* Interpretability and Explainability: Methods ensuring transparency and interpretability in FL models operating in complex environments.
* Personalization and Client Adaptivity: Techniques for tailoring FL models to individual clients while maintaining overall system robustness.
* Resource-Aware FL: Frameworks optimized for resource-limited devices, accounting for constraints such as memory, battery life, and connectivity.
* Application Case Studies: Demonstrations of FL’s performance in managing non-iid data and device heterogeneity in real-world applications.
* Tools and Benchmarks: Development of datasets and metrics reflecting the challenges posed by data and device variability.

Submission Information:
Deadline: January 30, 2025
Submission Process: Manuscripts should be submitted through the CMT paper submission website as a regular paper (Main Track), selecting the special session “Towards Robust Federated Learning: Addressing Data and Device Heterogeneity” as the primary subject area.
Review Process: All submissions will undergo the same rigorous review process as regular papers, with two reviewers from the special session program committee and a third reviewer assigned by the conference chairs.
Publication Proceedings: Accepted contributions will be published as part of the IJCNN 2025 conference proceedings.

We look forward to your contributions!

Special Session Organizers
Dr. Diletta Chiaro, University of Naples Federico II (Scholar)
Prof. Francesco Piccialli, University of Naples Federico II (Website; Scholar)
Dr. Fabio Giampaolo, University of Naples Federico II (Scholar)