Visible to the public Adaptive Laplace Mechanism: Differential Privacy Preservation in Deep Learning

TitleAdaptive Laplace Mechanism: Differential Privacy Preservation in Deep Learning
Publication TypeConference Paper
Year of Publication2017
AuthorsPhan, N., Wu, X., Hu, H., Dou, D.
Conference Name2017 IEEE International Conference on Data Mining (ICDM)
ISBN Number978-1-5386-3835-4
Keywordsadaptive Laplace mechanism, affine transformations, affine transforms, AI, artificial intelligence, Biological neural networks, Computational modeling, data privacy, Deep Learning, deep neural networks, Differential privacy, differential privacy preservation, Human Behavior, human factor, human factors, Laplace Mechanism, Laplace transforms, learning (artificial intelligence), loss functions, machine learning, neural nets, Neurons, privacy, privacy budget consumption, pubcrawl, resilience, Resiliency, Scalability, Training, training steps
Abstract

In this paper, we focus on developing a novel mechanism to preserve differential privacy in deep neural networks, such that: (1) The privacy budget consumption is totally independent of the number of training steps; (2) It has the ability to adaptively inject noise into features based on the contribution of each to the output; and (3) It could be applied in a variety of different deep neural networks. To achieve this, we figure out a way to perturb affine transformations of neurons, and loss functions used in deep neural networks. In addition, our mechanism intentionally adds "more noise" into features which are "less relevant" to the model output, and vice-versa. Our theoretical analysis further derives the sensitivities and error bounds of our mechanism. Rigorous experiments conducted on MNIST and CIFAR-10 datasets show that our mechanism is highly effective and outperforms existing solutions.

URLhttps://ieeexplore.ieee.org/document/8215511
DOI10.1109/ICDM.2017.48
Citation Keyphan_adaptive_2017