AutoPoly-HE: Autonomous Design of Polynomial Activations for Privacy-Preserving Encrypted Inference

본문 미리보기

arXiv:2610.19325v1 Announce Type: new Abstract: Privacy-preserving machine learning has become critical for deploying artificial intelligence in sensitive domains such as healthcare, finance, and infrastructure monitoring. Homomorphic encryption (HE) enables secure inference on encrypted data. Despite its advantages, HE is not compatible with nonlinear activation functions found in machine learning models. Existing methods mitigate this problem by manually designing polynomial approximations fo