Too Good to be True? Turn Any Model Differentially Private With DP-Weights

arXiv:2406.19507v1 Announce Type: cross
Abstract: Imagine training a machine learning model with Differentially Private Stochastic Gradient Descent (DP-SGD), only to discover post-training that the noise level was either too high, crippling your model's utility, or too low, compromising privacy. The dreaded realization hits: you must start the lengthy training process from scratch. But what if you could avoid this retraining nightmare? In this study, we introduce a groundbreaking approach (to our knowledge) that applies differential privacy noise to the model's …

arxiv cs.ai cs.cr cs.lg discover good high low machine machine learning noise privacy private start too good to be true training turn utility