Abstract: In today's world, many people interested to share the data, but they afraid about disclose of data so they want secure their data with communication so, the concept of privacy preservation of data is come in picture. Back-propagation is the most effective algorithm for training deep learning models. To protect the private data, the proposed model having the BGV encryption scheme to encrypt the private data and perform the high-order back propagation algorithm on the encrypted data. Sigmoid function as a polynomial function with the BGV encryption. The proposed algorithm helps to improve the efficiency of back-propagation. BGV is Homomorphic encryption technique that allows computations to be carried out on cipher text. Homomorphic encryption is the conversion of data into cipher text that can be help to analyse and work with as if it were still in its original form. Furthermore, the BGV encryption scheme is used to protect the private data during the learning process. Experiments show that our proposed scheme is secure and efficient.
Keywords: Privacy, Back-propagation, the BGV encryption scheme.