This paper attempts to answer the question whether neural network pruning can be used as a tool to achieve differential privacy without losing much data utility. As a first step towards understanding the relationship between neural network pruning and differential privacy, this paper proves that pruning a given layer of the neural network is equivalent to adding a certain amount of differentially private noise to its hidden-layer activations. The paper also presents experimental results to show the practical implications of the theoretical finding and the key parameter values in a simple practical setting. These results show that neural network pruning can be a more effective alternative to adding differentially private noise for neural networks.
@article{arxiv.2003.01876,
title = {Privacy-preserving Learning via Deep Net Pruning},
author = {Yangsibo Huang and Yushan Su and Sachin Ravi and Zhao Song and Sanjeev Arora and Kai Li},
journal= {arXiv preprint arXiv:2003.01876},
year = {2020}
}