Handling Spatial-Temporal Data Heterogeneity for Federated Continual Learning via Tail Anchor
This addresses data heterogeneity issues in federated continual learning for real-world applications, representing an incremental improvement with novel components.
The paper tackles the problem of spatial-temporal data heterogeneity in federated continual learning, which causes catastrophic forgetting, and proposes Federated Tail Anchor (FedTA) to adjust feature positions, achieving superior performance over existing methods while preserving feature relationships.
Federated continual learning (FCL) allows each client to continually update its knowledge from task streams, enhancing the applicability of federated learning in real-world scenarios. However, FCL needs to address not only spatial data heterogeneity between clients but also temporal data heterogeneity between tasks. In this paper, empirical experiments demonstrate that such input-level heterogeneity significantly affects the model's internal parameters and outputs, leading to severe spatial-temporal catastrophic forgetting of local and previous knowledge. To this end, we propose Federated Tail Anchor (FedTA) to mix trainable Tail Anchor with the frozen output features to adjust their position in the feature space, thereby overcoming parameter-forgetting and output-forgetting. Three novel components are also included: Input Enhancement for improving the performance of pre-trained models on downstream tasks; Selective Input Knowledge Fusion for fusion of heterogeneous local knowledge on the server; and Best Global Prototype Selection for finding the best anchor point for each class in the feature space. Extensive experiments demonstrate that FedTA not only outperforms existing FCL methods but also effectively preserves the relative positions of features.