Exact and Efficient Bayesian Inference for Privacy Risk Quantification (Extended Version)

• URL: http://arxiv.org/abs/2308.16700v1
• Date: Thu, 31 Aug 2023 13:04:04 GMT
• Title: Exact and Efficient Bayesian Inference for Privacy Risk Quantification (Extended Version)
• Authors: Rasmus C. Rønneberg, Raúl Pardo, Andrzej Wąsowski,
• Abstract summary: Privug is a method to quantify privacy risks of data analytics programs by analyzing their source code. The inference engine is implemented for a subset of Python programs. We evaluate the method by analyzing privacy risks in programs to release public statistics.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Data analysis has high value both for commercial and research purposes. However, disclosing analysis results may pose severe privacy risk to individuals. Privug is a method to quantify privacy risks of data analytics programs by analyzing their source code. The method uses probability distributions to model attacker knowledge and Bayesian inference to update said knowledge based on observable outputs. Currently, Privug uses Markov Chain Monte Carlo (MCMC) to perform inference, which is a flexible but approximate solution. This paper presents an exact Bayesian inference engine based on multivariate Gaussian distributions to accurately and efficiently quantify privacy risks. The inference engine is implemented for a subset of Python programs that can be modeled as multivariate Gaussian models. We evaluate the method by analyzing privacy risks in programs to release public statistics. The evaluation shows that our method accurately and efficiently analyzes privacy risks, and outperforms existing methods. Furthermore, we demonstrate the use of our engine to analyze the effect of differential privacy in public statistics.

Related papers

• Pseudo-Probability Unlearning: Towards Efficient and Privacy-Preserving Machine Unlearning [59.29849532966454]
  We propose PseudoProbability Unlearning (PPU), a novel method that enables models to forget data to adhere to privacy-preserving manner. Our method achieves over 20% improvements in forgetting error compared to the state-of-the-art.
  arXiv  Detail & Related papers  (2024-11-04T21:27:06Z)
• Initialization Matters: Privacy-Utility Analysis of Overparameterized Neural Networks [72.51255282371805]
  We prove a privacy bound for the KL divergence between model distributions on worst-case neighboring datasets. We find that this KL privacy bound is largely determined by the expected squared gradient norm relative to model parameters during training.
  arXiv  Detail & Related papers  (2023-10-31T16:13:22Z)
• Conditional Density Estimations from Privacy-Protected Data [0.0]
  We propose simulation-based inference methods from privacy-protected datasets. We illustrate our methods on discrete time-series data under an infectious disease model and with ordinary linear regression models.
  arXiv  Detail & Related papers  (2023-10-19T14:34:17Z)
• On the Properties and Estimation of Pointwise Mutual Information Profiles [49.877314063833296]
  The pointwise mutual information profile, or simply profile, is the distribution of pointwise mutual information for a given pair of random variables. We introduce a novel family of distributions, Bend and Mix Models, for which the profile can be accurately estimated using Monte Carlo methods.
  arXiv  Detail & Related papers  (2023-10-16T10:02:24Z)
• PrivacyMind: Large Language Models Can Be Contextual Privacy Protection Learners [81.571305826793]
  We introduce Contextual Privacy Protection Language Models (PrivacyMind) Our work offers a theoretical analysis for model design and benchmarks various techniques. In particular, instruction tuning with both positive and negative examples stands out as a promising method.
  arXiv  Detail & Related papers  (2023-10-03T22:37:01Z)
• Differentially Private Linear Regression with Linked Data [3.9325957466009203]
  Differential privacy, a mathematical notion from computer science, is a rising tool offering robust privacy guarantees. Recent work focuses on developing differentially private versions of individual statistical and machine learning tasks. We present two differentially private algorithms for linear regression with linked data.
  arXiv  Detail & Related papers  (2023-08-01T21:00:19Z)
• Data Analytics with Differential Privacy [0.0]
  We develop differentially private algorithms to analyze distributed and streaming data. In the distributed model, we consider the particular problem of learning -- in a distributed fashion -- a global model of the data. We offer one of the strongest privacy guarantees for the streaming model, user-level pan-privacy.
  arXiv  Detail & Related papers  (2023-07-20T17:43:29Z)
• Simulation-based, Finite-sample Inference for Privatized Data [14.218697973204065]
  We propose a simulation-based "repro sample" approach to produce statistically valid confidence intervals and hypothesis tests. We show that this methodology is applicable to a wide variety of private inference problems.
  arXiv  Detail & Related papers  (2023-03-09T15:19:31Z)
• A General Framework for Auditing Differentially Private Machine Learning [27.99806936918949]
  We present a framework to statistically audit the privacy guarantee conferred by a differentially private machine learner in practice. Our work develops a general methodology to empirically evaluate the privacy of differentially private machine learning implementations.
  arXiv  Detail & Related papers  (2022-10-16T21:34:18Z)
• Leveraging Unlabeled Data to Predict Out-of-Distribution Performance [63.740181251997306]
  Real-world machine learning deployments are characterized by mismatches between the source (training) and target (test) distributions. In this work, we investigate methods for predicting the target domain accuracy using only labeled source data and unlabeled target data. We propose Average Thresholded Confidence (ATC), a practical method that learns a threshold on the model's confidence, predicting accuracy as the fraction of unlabeled examples.
  arXiv  Detail & Related papers  (2022-01-11T23:01:12Z)
• Differentially Private Federated Learning with Laplacian Smoothing [72.85272874099644]
  Federated learning aims to protect data privacy by collaboratively learning a model without sharing private data among users. An adversary may still be able to infer the private training data by attacking the released model. Differential privacy provides a statistical protection against such attacks at the price of significantly degrading the accuracy or utility of the trained models.
  arXiv  Detail & Related papers  (2020-05-01T04:28:38Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.