Encrypted Neural Networks without Overflows

ArXi:2605.23096v1 Announce Type: cross Fully homomorphic encryption (FHE) enables private inference by evaluating neural networks on encrypted data. In this way, we can delegate the computation to a third party server without ever revealing the user's data. Currently, the CKKS scheme is the backbone of most efficient FHE implementations, but it only s addition, multiplication, and array rotation operations, thus requiring all activation functions of the neural network to be approximated by polynomials within a certain interval, imposing strict design tolerances.