FedProK: Trustworthy Federated Class-Incremental Learning via Prototypical Feature Knowledge Transfer

arXiv:2405.02685v1 Announce Type: new
Abstract: Federated Class-Incremental Learning (FCIL) focuses on continually transferring the previous knowledge to learn new classes in dynamic Federated Learning (FL). However, existing methods do not consider the trustworthiness of FCIL, i.e., improving continual utility, privacy, and efficiency simultaneously, which is greatly influenced by catastrophic forgetting and data heterogeneity among clients. To address this issue, we propose FedProK (Federated Prototypical Feature Knowledge Transfer), leveraging prototypical feature as a novel representation of knowledge to perform spatial-temporal knowledge …

abstract arxiv catastrophic forgetting class continual cs.ai cs.lg cs.ne dynamic efficiency feature federated learning however improving incremental knowledge learn privacy transfer trustworthy type utility via