FedKDD 2025: International Joint Workshop on Federated Learning for Data Mining and Graph Analytics

Goals

Recent advances in deep learning, especially the rise of large language models, have greatly impacted fields such as urban planning and healthcare. Data now come from many sources, including structured tabular records and graphs that capture complex relationships between samples. This diversity has increased the need for scalable and privacy-preserving learning methods. Federated Learning (FL) is a powerful tool for aggregating knowledge from distributed data while protecting privacy. However, applying FL to new areas like large-scale language modeling, multi-agent systems, and AI-driven scientific discovery requires new methods, as these fields face challenges in scalability, trustworthiness, and robustness. This workshop aims to encourage discussion on distributed data mining and graph analytics in this evolving technological landscape, and it invites researchers and practitioners to share innovative algorithms, system designs, applications, and evaluation strategies that address these challenges and drive the development of trustworthy intelligent systems.