Skip to content
arXiv· arXiv:2103.08987· Sylvain Chatel, Apostolos Pyrgelis, Juan Ramon Troncoso-Pastoriza, Jean-Pierre Hubaux·Indexed by this site · Originally published

SoK: Privacy-Preserving Collaborative Tree Model Learning

SoK: Privacy-Preserving Collaborative Tree-based Model Learning

AI summary

A SoK study systematically reviews the literature on distributed privacy-preserving training of tree models along four dimensions: learning algorithm, collaboration model, protection mechanism, and threat model. Based on this, the authors identify the strengths and limitations of each scheme, and state that they are the first to propose a framework for analyzing information leakage in distributed tree model training.

Read originalProceedings on Privacy Enhancing Technologies