Privacy-preserving feature selection with Secure Multiparty Computation

Xiling Li, Rafael Baiao Dowsley, Martine de Cock

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review


Existing work on privacy-preserving machine learning with Secure Multiparty Computation (MPC) is almost exclusively focused on model training and on inference with trained models, thereby overlooking the important data pre-processing stage. In this work, we propose the first MPC based protocol for private feature selection based on the filter method, which is independent of model training, and can be used in combination with any MPC protocol to rank features. We propose an efficient feature scoring protocol based on Gini impurity to this end. To demonstrate the feasibility of our approach for practical data science, we perform experiments with the proposed MPC protocols for feature selection in a commonly used machine-learning-as-a-service configuration where computations are outsourced to multiple servers, with semi-honest and with malicious adversaries. Regarding effectiveness, we show that secure feature selection with the proposed protocols improves the accuracy of classifiers on a variety of real-world data sets, without leaking information about the feature values or even which features were selected. Regarding efficiency, we document runtimes ranging from several seconds to an hour for our protocols to finish, depending on the size of the data set and the security settings.
Original languageEnglish
Title of host publicationProceedings of the 38th International Conference on Machine Learning
EditorsMarina Meila, Tong Zhang
Place of PublicationLondon UK
PublisherProceedings of Machine Learning Research (PMLR)
Number of pages11
Publication statusPublished - 2021
EventInternational Conference on Machine Learning 2021 - Online, United States of America
Duration: 18 Jul 202124 Jul 2021
Conference number: 38th (Proceedings) (Website)


ConferenceInternational Conference on Machine Learning 2021
Abbreviated titleICML 2021
Country/TerritoryUnited States of America
Internet address

Cite this