Efficient-FedRec: Efficient Federated Learning Framework for Privacy-Preserving News Recommendation

• DOI: 10.18653/v1/2021.emnlp-main.223
• Publication Date: 2021-12-17T03:56:42Z
• Authors: Jingwei Yi; Fangzhao Wu; Chuhan Wu; Ruixuan Liu; Guangzhong Sun; Xing Xie
• Keywords: Federated Learning; Private information retrieval

News recommendation is critical for personalized news access. Most existing methods rely on centralized storage of users’ historical click behavior data, which may lead to privacy concerns and hazards. Federated Learning a privacy-preserving framework multiple clients collaboratively train models without sharing their private data. However, the computation communication cost directly learning many in federated way are unacceptable user clients. In this paper, we propose an efficient recommendation. Instead training communicating whole model, decompose model into large maintained server light-weight shared both clients, where representations communicated between More specifically, request from server, send locally computed gradients aggregation. The updates its global with aggregated gradients, further infer updated representations. Since local contain information, secure aggregation method aggregate way. Experiments two real-world datasets show that our can reduce while keep promising performance.