【论文记录】Differentially Private Empirical Risk Minimization with Input Perturbation

Problem Definition and Preliminary

• privacy concerns
  data privacy : when the data contributors release their own data to the database
  model privacy : when the database publishes the learned model to the model user
• 采用的方法
  针对 model privacy : ($\epsilon$,$\delta$)-differential privacy
  针对 data privacy : ($\epsilon$,$\delta$)-local differential privacy     \,\,\, (注 : 此种定义通常用在 under the non-interactive setting)
• how to assess utility
  借助 empirical excess risk     \,\,\, the empirical excess risk of $\mathcal{M}$ that randomly outputs $w$ is defined as   \, $J(w;D)-J(w^{\ast };D)$
  
  
  

Ref

Fukuchi, K., Tran, Q. K., & Sakuma, J. (2017, October). Differentially private empirical risk minimization with input perturbation. In International Conference on Discovery Science (pp. 82-90). Springer, Cham.