Dynamic Differential Privacy Preserving Algorithm Based on Deep Learning
There may be some sensitive information of users in the training dataset used in deep learning,and during the model training process,these sensitive information will implicitly exist in the model parameters,thus there is a risk of priva-cy leakage.In this paper,we introduce differential privacy of Gaussian mechanism on the basis of AdamP optimizer,and propose a dynamic privacy budget allocation algorithm based on primary power function to allocate the privacy budget of dif-ferential privacy more reasonably,i.e.DP-AdamP,so as to better balance the privacy and model accuracy.Experimental re-sults show that the DP-AdamP algorithm in this paper has better accuracy than the traditional DP-SGD algorithm under the same privacy budget,about 7.7％higher in the case of low privacy budget,and about 3.9％higher in the case of high priva-cy budget,and separate experiments are conducted for the more practically relevant case of low privacy budget,which fur-ther validates the effectiveness of the DP-AdamP algorithm.
deep learningdifferential privacyadaptive privacy budgetAdamP algorithm
国家科技期刊平台
NSTL
万方数据
