English

Preventing Adversarial Use of Datasets through Fair Core-Set Construction

Machine Learning 2019-10-25 v1 Artificial Intelligence Machine Learning

Abstract

We propose improving the privacy properties of a dataset by publishing only a strategically chosen "core-set" of the data containing a subset of the instances. The core-set allows strong performance on primary tasks, but forces poor performance on unwanted tasks. We give methods for both linear models and neural networks and demonstrate their efficacy on data.

Keywords

Cite

@article{arxiv.1910.10871,
  title  = {Preventing Adversarial Use of Datasets through Fair Core-Set Construction},
  author = {Benjamin Spector and Ravi Kumar and Andrew Tomkins},
  journal= {arXiv preprint arXiv:1910.10871},
  year   = {2019}
}

Comments

6 pages, 2 figures, NeurIPS 2019 Privacy In Machine Learning Workshop (PriML 2019)

R2 v1 2026-06-23T11:53:14.705Z