Corpus ID: 231583084

Adversary Instantiation: Lower Bounds for Differentially Private Machine Learning

@article{Nasr2021AdversaryIL,
  title={Adversary Instantiation: Lower Bounds for Differentially Private Machine Learning},
  author={Milad Nasr and Shuang Song and Abhradeep Thakurta and Nicolas Papernot and Nicholas Carlini},
  journal={ArXiv},
  year={2021},
  volume={abs/2101.04535}
}

• Milad Nasr, Shuang Song, +2 authors Nicholas Carlini
• Published 2021
• Computer Science
• ArXiv

Differentially private (DP) machine learning allows us to train models on private data while limiting data leakage. DP formalizes this data leakage through a cryptographic game, where an adversary must predict if a model was trained on a dataset D, or a dataset D′ that differs in just one example. If observing the training algorithm does not meaningfully increase the adversary’s odds of successfully guessing which dataset the model was trained on, then the algorithm is said to be differentially… Expand

View PDF on arXiv

Save to Library

Create Alert

Cite

Launch Research Feed