LESS-VFL: 用于联合学习的高效通信特征选择（英文版）

英文标题：LESS-VFL: Communication-Efficient Feature Selection for Vertical Federated Learning中文摘要：本文提出了 LESS-VFL，这是一种用于垂直切分数据的分布式系统中的通信高效特征选择方法，通过优化服务器端的全局模型，然后将此信息分享给其他参与方，从而允许本地特征选择，进而提高模型的泛化性、效率和可解释性。英文摘要：We propose LESS-VFL, a communication-efficient feature selection method fordistributed systems with vertically partitioned data. We consider a system of aserver and several parties with local datasets that share a sample ID space buthave different feature sets. The parties wish to collaborativel