SGD on Random Mixtures - Private Machine Learning under Data Breach Threats.
ICLR(2018)
摘要
We propose Stochastic Gradient Descent on Random Mixtures (SGDRM) as a simple way of protecting data under data breach threats. We show that SGDRM converges to the globally optimal point for deep neural networks with linear activations while being differentially private. We also train nonlinear neural networks with private mixtures as the training data, proving the practicality of SGDRM.
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络