A Secure Aggregation for Federated Learning on Long-Tailed Data
This addresses security and data imbalance issues in Federated Learning, but it appears incremental as it builds on existing aggregation methods.
The paper tackles the challenges of unbalanced long-tailed data distribution and Byzantine attacks in Federated Learning by proposing a novel two-layer aggregation method that rejects malicious models and selects valuable ones with tail class information, validated through preliminary experiments.
As a distributed learning, Federated Learning (FL) faces two challenges: the unbalanced distribution of training data among participants, and the model attack by Byzantine nodes. In this paper, we consider the long-tailed distribution with the presence of Byzantine nodes in the FL scenario. A novel two-layer aggregation method is proposed for the rejection of malicious models and the advisable selection of valuable models containing tail class data information. We introduce the concept of think tank to leverage the wisdom of all participants. Preliminary experiments validate that the think tank can make effective model selections for global aggregation.