Title: Privacy-Preserving Tree-Based Inference with TFHE

Abstract: Privacy enhancing technologies (PETs) have been proposed as a way to protect the privacy of data while still allowing for data analysis. In this work, we focus on Fully Homomorphic Encryption (FHE), a powerful tool that allows for arbitrary computations to be performed on encrypted data. FHE has received lots of attention in the past few years and has reached realistic execution times and correctness.
More precisely, we explain in this paper how we apply FHE to tree-based models and get state-of-the-art solutions over encrypted tabular data. We show that our method is applicable to a wide range of tree-based models, including decision trees, random forests, and gradient boosted trees, and has been implemented within the Concrete-ML library, which is open-source at this https URL With a selected set of use-cases, we demonstrate that our FHE version is very close to the unprotected version in terms of accuracy.