DarKnight: A Data Privacy Scheme for Training and Inference of Deep Neural Networks
arXiv:2006.01300
Abstract
Protecting the privacy of input data is of growing importance as machine learning methods reach new application domains. In this paper, we provide a unified training and inference framework for large DNNs while protecting input privacy and computation integrity. Our approach called DarKnight uses a novel data blinding strategy using matrix masking to create input obfuscation within a trusted execution environment (TEE). Our rigorous mathematical proof demonstrates that our blinding process provides information-theoretic privacy guarantee by bounding information leakage. The obfuscated data can then be offloaded to any GPU for accelerating linear operations on blinded data. The results from linear operations on blinded data are decoded before performing non-linear operations within the TEE. This cooperative execution allows DarKnight to exploit the computational power of GPUs to perform linear operations while exploiting TEEs to protect input privacy. We implement DarKnight on an Intel SGX TEE augmented with a GPU to evaluate its performance.
References in corpus (16)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
- Deep Learning with Differential Privacy
- DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradients
- Learning to Communicate with Deep Multi-Agent Reinforcement Learning
- An Analysis of Deep Neural Network Models for Practical Applications
- Slalom: Fast, Verifiable and Private Execution of Neural Networks in Trusted Hardware
- Chiron: Privacy-preserving Machine Learning as a Service
- Rethinking floating point for deep learning
- Efficient Deep Learning on Multi-Source Private Data
- A Study of BFLOAT16 for Deep Learning Training
- CodedPrivateML: A Fast and Privacy-Preserving Framework for Distributed Machine Learning
- Privacy-Preserving Inference in Machine Learning Services Using Trusted Execution Environments
- Attacking Binarized Neural Networks
- Privacy Aware Offloading of Deep Neural Networks
- Secure multiparty computations in floating-point arithmetic