analiza aiqbench
This research advances the intersection of machine learning and cryptography by establishing rigorous differentially private permutation tests that maintain statistical validity without sacrificing efficiency. By introducing kernel-based methods like dpMMD and dpHSIC, the work provides a practical framework for hypothesis testing on sensitive data while achieving minimax optimal power across various privacy regimes. The availability of public code further facilitates adoption in real-world scenarios requiring robust privacy guarantees alongside standard analytical rigor.
Differentially Private Permutation Tests
arXiv:2310.19043v3 Announce Type: replace-cross Abstract: Recent years have witnessed growing concerns about the privacy of sensitive data. In response to these concerns, differential privacy has emerged as a rigorous framework for privacy protection, gaining widespread recognition in both academic and industrial circles. While substantial progress has been made in private data analysis, existing methods often suffer from impracticality or a significant loss of statistical efficiency. This paper aims to alleviate these concerns in the context of hypothesis testing by introducing differentially private permutation tests. The proposed framework extends classical non-private permutation tests to private settings, maintaining both finite-sample validity and differential privacy in a rigorous manner. The power of the proposed test depends on the choice of a test statistic, and we establish general conditions for consistency and non-asymptotic uniform power. To demonstrate the utility and practicality of our framework, we focus on reproducing kernel-based test statistics and introduce differentially private kernel tests for two-sample and independence testing: dpMMD and dpHSIC. The proposed kernel tests are straightforward to implement, applicable to various types of data, and attain minimax optimal power across different privacy regimes. Our empirical evaluations further highlight their competitive power under various synthetic and real-world scenarios, emphasizing their practical value. The code is publicly available to facilitate the implementation of our framework.
Abstract:Recent years have witnessed growing concerns about the privacy of sensitive data. In response to these concerns, differential privacy has emerged as a rigorous framework for privacy protection, gaining widespread recognition in both academic and industrial circles. While substantial progress has been made in private data analysis, existing methods often suffer from impracticality or a significant loss of statistical efficiency. This paper aims to alleviate these concerns in the context of hypothesis testing by introducing differentially private permutation tests. The proposed framework extends classical non-private permutation tests to private settings, maintaining both finite-sample validity and differential privacy in a rigorous manner. The power of the proposed test depends on the choice of a test statistic, and we establish general conditions for consistency and non-asymptotic uniform power. To demonstrate the utility and practicality of our framework, we focus on reproducing kernel-based test statistics and introduce differentially private kernel tests for two-sample and independence testing: dpMMD and dpHSIC. The proposed kernel tests are straightforward to implement, applicable to various types of data, and attain minimax optimal power across different privacy regimes. Our empirical evaluations further highlight their competitive power under various synthetic and real-world scenarios, emphasizing their practical value. The code is publicly available to facilitate the implementation of our framework.
Statistics Theory (math.ST); Cryptography and Security (cs.CR); Machine Learning (cs.LG); Methodology (stat.ME); Machine Learning (stat.ML)
arXiv:2310.19043 [math.ST]
(or arXiv:2310.19043v3 [math.ST] for this version)
Text integral sub licența deschisă a publicației · 232 cuvinte
Subiecte conexe
Differentially Private Joint Independence Test
arXiv · 5 iun. 2026
Differentially Private Verification of Distribution Properties
arXiv · 19 aug. 2026
Research Note: Filtering Subversion-Relevant Information From Pretraining Data Is Feasible
LessWrong · 5 oct. 2026
An Adaptive Gradient Clipping and Noise Injection Mechanism for Differentially Private Federated Learning
arXiv · 19 aug. 2026