Federated continual learning: A comprehensive survey on lifelong and privacy-preserving learning over distributed and non-stationary data

The article presents a comprehensive survey on Federated Continual Learning (FCL), which integrates principles of Federated Learning (FL) and Continual Learning (CL) to address challenges posed by non-stationary data in distributed settings. It defines the FCL problem, analyzes the limitations of traditional FL in dynamic environments, and proposes a taxonomy of FCL approaches while reviewing applications and evaluation metrics. This work is significant for practitioners as it identifies key challenges and opportunities for developing scalable, privacy-preserving systems that can adapt to evolving data distributions in real-world applications.