Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
Gemini 2.5 Pro
GPT-5
GPT-4o
DeepSeek R1 via Azure
2000 character limit reached

Declarative Privacy-Preserving Inference Queries (2401.12393v3)

Published 22 Jan 2024 in cs.DB and cs.AI

Abstract: Detecting inference queries running over personal attributes and protecting such queries from leaking individual information requires tremendous effort from practitioners. To tackle this problem, we propose an end-to-end workflow for automating privacy-preserving inference queries including the detection of subqueries that involve AI/ML model inferences on sensitive attributes. Our proposed novel declarative privacy-preserving workflow allows users to specify "what private information to protect" rather than "how to protect". Under the hood, the system automatically chooses privacy-preserving plans and hyper-parameters.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (13)
  1. Deep learning with differential privacy. In Proceedings of the 2016 ACM SIGSAC conference on computer and communications security. 308–318.
  2. On the opportunities and risks of foundation models. arXiv preprint arXiv:2108.07258 (2021).
  3. Christine L Borgman and Amy Brand. 2022. Data blind: Universities lag in capturing and exploiting data. Science 378, 6626 (2022), 1278–1281.
  4. The algorithmic foundations of differential privacy. Foundations and Trends® in Theoretical Computer Science 9, 3–4 (2014), 211–407.
  5. Model inversion attacks that exploit confidence information and basic countermeasures. In Proceedings of the 22nd ACM SIGSAC conference on computer and communications security. 1322–1333.
  6. Erica Jain. 2022. Evolving Consumer Expectations Are Driving Healthcare Innovation—For The Better. Forbes (2022). https://www.forbes.com/sites/forbesbusinesscouncil/2023/02/17/evolving-consumer-expectations-are-driving-healthcare-innovation-for-the-better/?sh=4dda9c283891
  7. Membership inference attacks against machine learning models. In 2017 IEEE symposium on security and privacy (SP). IEEE, 3–18.
  8. Fbnetv2: Differentiable neural architecture search for spatial and channel dimensions. In CVPR.
  9. Differentially private fine-tuning of language models. arXiv preprint arXiv:2110.06500 (2021).
  10. A neural database for differentially private spatial range queries. arXiv preprint arXiv:2108.01496 (2021).
  11. DeepMapping: The Case for Learned Data Mapping for Compression and Efficient Query Processing. arXiv preprint arXiv:2307.05861 (2023).
  12. Benchmark of DNN Model Search at Deployment Time. In Proceedings of the 34th International Conference on Scientific and Statistical Database Management.
  13. Privacy-Preserving Redaction of Diagnosis Data through Source Code Analysis. In Proceedings of the 35th International Conference on Scientific and Statistical Database Management. 1–4.

Summary

We haven't generated a summary for this paper yet.

Dice Question Streamline Icon: https://streamlinehq.com

Follow-up Questions

We haven't generated follow-up questions for this paper yet.

X Twitter Logo Streamline Icon: https://streamlinehq.com
Youtube Logo Streamline Icon: https://streamlinehq.com