Memetic Federated Learning for Biomedical Natural Language Processing

Xinya Zhou, Conghui Tan, Di Jiang, Bosen Zhang, Si Li, Yajing Xu, Qian Xu, Sheng Gao

Abstract: Privacy protection is an essential issue in biomedical natural language processing (BioNLP). Recently, some researchers apply federated learning (FL) in BioNLP to protect the privacy of biomedical data. However, their methods are only applicable for small NLP models, whose effectiveness is heavily limited in processing biomedical data. In this paper, we propose a novel memetic federated learning framework named Mem-Fed, which is tailored for federated learning of large-scale NLP models in the biomedical scenario. Experiments with large-scale BioNLP model on the public dataset show that the proposed framework significantly outperforms the state-of-the-art counterparts.