Secure Collaborative Training and Inference for XGBoost

Autor:	Law, Andrew, Leung, Chester, Poddar, Rishabh, Popa, Raluca Ada, Shi, Chenyu, Sima, Octavian, Yu, Chaofan, Zhang, Xingmeng, Zheng, Wenting
Rok vydání:	2020
Předmět:	Computer Science - Cryptography and Security
Druh dokumentu:	Working Paper
Popis:	In recent years, gradient boosted decision tree learning has proven to be an effective method of training robust models. Moreover, collaborative learning among multiple parties has the potential to greatly benefit all parties involved, but organizations have also encountered obstacles in sharing sensitive data due to business, regulatory, and liability concerns. We propose Secure XGBoost, a privacy-preserving system that enables multiparty training and inference of XGBoost models. Secure XGBoost protects the privacy of each party's data as well as the integrity of the computation with the help of hardware enclaves. Crucially, Secure XGBoost augments the security of the enclaves using novel data-oblivious algorithms that prevent access side-channel attacks on enclaves induced via access pattern leakage. Comment: ACM CCS PPMLP'20
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2010.02524 Zobrazit plný text záznamu View this record from Arxiv