Explore federated learning schemes

Background: Federated learning (FL) offers a pathway for privacy-preserving model sharing across distributed monitoring networks, though its application in this context is open.

Question / Future Work: Explore the potential of applying federated learning schemes to air quality forecasting models to enable privacy-preserving model sharing across distributed monitoring networks, which could concurrently improve generalization across diverse urban environments.

Why It Matters: This addresses the dual challenge of data privacy constraints and the need for models generalized across multiple urban areas.

Evidence: Another important direction is the exploration of federated learning schemes, enabling privacy-preserving model sharing across distributed monitoring networks while improving generalization across diverse urban environments.