English

Regression with Label Differential Privacy

Machine Learning 2023-10-06 v3 Cryptography and Security

Abstract

We study the task of training regression models with the guarantee of label differential privacy (DP). Based on a global prior distribution on label values, which could be obtained privately, we derive a label DP randomization mechanism that is optimal under a given regression loss function. We prove that the optimal mechanism takes the form of a "randomized response on bins", and propose an efficient algorithm for finding the optimal bin values. We carry out a thorough experimental evaluation on several datasets demonstrating the efficacy of our algorithm.

Keywords

Cite

@article{arxiv.2212.06074,
  title  = {Regression with Label Differential Privacy},
  author = {Badih Ghazi and Pritish Kamath and Ravi Kumar and Ethan Leeman and Pasin Manurangsi and Avinash V Varadarajan and Chiyuan Zhang},
  journal= {arXiv preprint arXiv:2212.06074},
  year   = {2023}
}

Comments

Appeared at ICLR '23, 28 pages, 6 figures

R2 v1 2026-06-28T07:31:30.189Z