Global ETD Search

Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.

Results

Showing 1 to 20 of 52 for “"Membership Inference"”.

  1. Using Approximate Bayesian Neural Networks for Membership Inference

    … settings, evaluating their vulnerability to Membership Inference Attacks (MIA) has become an important security concern. These attacks aim to determine whether a given data point was part of a target model’s training set, thereby potentially revealing sensitive information about individuals …

    chalmers Repository record for Using Approximate Bayesian Neural Networks for Membership Inference (opens in a new tab)

  2. Honesty Is Not Always the Best Policy: Defending Against Membership Inference Attacks on Genomic Data

    … public. Several of these concerns revolve around membership-inference attacks, whereby an adversarial actor attempts to infer a target individual's presence in a given dataset. Membership-inference attacks pose a significant privacy risk, as genomic datasets often contain potentially sensitive …

    wustl Repository record for Honesty Is Not Always the Best Policy: Defending Against Membership Inference Attacks on Genomic Data (opens in a new tab)

  3. Privacy Risks in Time Series Models: Membership Inference in Deep Learning-Based Time Series Forecasting Models

    … to leak sensitive data, for example through membership inference attacks (MIAs), raises serious privacy concerns. While MIAs are well-studied in domains like image classification, their implications for time series models remain underexplored. This thesis investigates the vulnerability of …

    chalmers Repository record for Privacy Risks in Time Series Models: Membership Inference in Deep Learning-Based Time Series Forecasting Models (opens in a new tab)

  4. Semantically Aware Attacks on Text-based Models: An Extension of Context-aware and Neighbourhood Comparisonbased Membership Inference Attacks

    … can be extracted from the trained models. In a membership inference attack (MIA), an adversary is expected to have access to a trained model θ and a data sample d, sampled from the same distribution as the unknown training data. The objective of the adversary is to construct an algorithm A(θ, d) …

    chalmers Repository record for Semantically Aware Attacks on Text-based Models: An Extension of Context-aware and Neighbourhood Comparisonbased Membership Inference Attacks (opens in a new tab)

  5. Data Centric Defenses for Privacy Attacks

    … are implemented for two attacks namely membership inference and model inversion using two distinct techniques. While the proposed augmentations offer a better privacy-utility tradeoff on CIFAR-10 for membership inference, they reduce the reconstruction rate to ≤ 1% while reducing the …

    vt Repository record for Data Centric Defenses for Privacy Attacks (opens in a new tab)

  6. TOWARD A BETTER UNDERSTANDING OF PRIVACY LEAKAGE IN MACHINE LEARNING, USING DATASET PRUNING ATTACK

    While membership inference attacks in machine learning setups have been studied for several years, the exact mechanisms leading to privacy leakage are only partially understood. We developed a dataset pruning attack, which is similar to a dataset poisoning attack, but that differs in that it …

    nus Repository record for TOWARD A BETTER UNDERSTANDING OF PRIVACY LEAKAGE IN MACHINE LEARNING, USING DATASET PRUNING ATTACK (opens in a new tab) Download file for TOWARD A BETTER UNDERSTANDING OF PRIVACY LEAKAGE IN MACHINE LEARNING, USING DATASET PRUNING ATTACK (opens in a new tab)

  7. Is This Data Point In your Training Set? Similarity-based Inference Attacks: Performance Evaluation of Range Membership Attacks to Audit Privacy Risks in Machine Learning Models

    Membership Inference Attacks (MIAs) pose a serious threat to the privacy of machine learning (ML) models by determining whether a specific data point was used during model training. A recent and powerful variant, the Range Membership Inference Attack (RaMIA), assesses privacy risks over a range of …

    chalmers Repository record for Is This Data Point In your Training Set? Similarity-based Inference Attacks: Performance Evaluation of Range Membership Attacks to Audit Privacy Risks in Machine Learning Models (opens in a new tab)

  8. Data Privacy in the Modern Machine Learning Ecosystem

    … new vulnerabilities and privacy risks, such as membership inference attacks, attribute inference attacks, and data poisoning attacks. Such attacks can allow for the malicious manipulation of model outcomes and can cause serious violations to the data privacy of those individuals who have …

    gatech Repository record for Data Privacy in the Modern Machine Learning Ecosystem (opens in a new tab)

  9. Privacy and Robustness of Deep Neural Networks

    … discuss ap- proaches to defending models against membership inference attacks and adversarial attacks. Membership inference attacks attempt to infer the training set information from observing the model output, while adversarial attacks aim to alter the output of models via introducing minimal …

    duke Repository record for Privacy and Robustness of Deep Neural Networks (opens in a new tab)

  10. On Identifying and Mitigating Against Vulnerabilities of Machine Learning Models

    … the privacy vulnerability of ML models using membership inference (MI) attacks to facilitate the design and training of privacy- preserving models. An MI attack is a fundamental attack that has been widely considered the simplest attack to evaluate the privacy of the training data of ML models …

    auckland-ms Repository record for On Identifying and Mitigating Against Vulnerabilities of Machine Learning Models (opens in a new tab)

  11. Privacy Protection via Adversarial Examples

    … by attackers to perform automated, large-scale inference attacks. For instance, in attribute inference attacks, an attacker can use a machine learning classifier to predict a target user's private, sensitive attributes (e.g., gender, political view) via the public data shared by the user (e.g., …

    duke Repository record for Privacy Protection via Adversarial Examples (opens in a new tab)

  12. PANORAMIA : privacy auditing of machine learning models without retraining

    … for machine learning models that relies on membership inference attacks using generated data as non-members. By relying on generated non-member data, PANORAMIA eliminates the common dependency of privacy measurement tools on in-distribution non-member data. As a result, PANORAMIA does not …

    ubc Repository record for PANORAMIA : privacy auditing of machine learning models without retraining (opens in a new tab) Download PDF for PANORAMIA : privacy auditing of machine learning models without retraining (opens in a new tab)

  13. Emerging machine learning methods and their applications to life sciences

    … risks and strengthening resilience against membership inference threats. First, we study early dropout prediction for healthy ageing applications using a real dataset characterized by non-IID client distributions and severe class imbalance. We propose and evaluate principled client-selection …

    greece Repository record for Emerging machine learning methods and their applications to life sciences (opens in a new tab)

  14. PRIVACY-PRESERVING DATA GENERATION AND REMOVAL IN GENERATIVE MODELS

    … Differential privacy and privacy auditing via membership inference attacks are further incorporated, and the framework is extended to time-series data through VFLGAN-TS, which captures both temporal and attribute dependencies. For data deletion in diffusion models, the dissertation identifies …

    nus Repository record for PRIVACY-PRESERVING DATA GENERATION AND REMOVAL IN GENERATIVE MODELS (opens in a new tab) Download file for PRIVACY-PRESERVING DATA GENERATION AND REMOVAL IN GENERATIVE MODELS (opens in a new tab)

  15. Additive logistic mechanism for privacy-preserving self-supervised learning

    … alleviating the need for retraining. We apply membership inference attacks on both unprotected and protected models to quantify the trade-off between the model’s privacy and performance. We prove that the post-training protection algorithm is differentially private and empirically show that …

    texas Repository record for Additive logistic mechanism for privacy-preserving self-supervised learning (opens in a new tab)

  16. Approaching Memorization in Large Language Models

    … in LLMs by adopting a pipeline that combines membership inference and data extraction attacks, and we evaluate memorization across multiple models. Through systematic experiments, we analyze how memorization varies with model size, architecture, and content category. We observe memorization …

    waterloo Repository record for Approaching Memorization in Large Language Models (opens in a new tab)

  17. Privacy Risks in Text Masking Models for Anonymization

    … models by evaluating their vulnerability to Membership Inference Attacks (MIAs) and extraction attacks. MIAs are attempting to identify whether or not a data point was part of the training dataset, knowledge of the membership can in certain scenarios be a breach of privacy. Two …

    chalmers Repository record for Privacy Risks in Text Masking Models for Anonymization (opens in a new tab)

  18. Evaluating Privacy Metrics for Synthetic Tabular Data

    … We also introduce the Step-wise Error Comparing Membership Inference (SECMI) attack, which assesses prediction errors at each generation step to infer membership status. The study concludes that diffusion models, such as TabDDPM, generally achieve a superior balance of utility and privacy …

    waterloo Repository record for Evaluating Privacy Metrics for Synthetic Tabular Data (opens in a new tab)

  19. Privacy-preserving synthetic image data generation and classification

    … privacy is crucial due to its vulnerability to “membership inference attacks.” State-of-the-art (e.g., differential privacy, etc.) defenses compromised task accuracy to preserve privacy, and some methods reuse private data or require more public data, which is impractical in some domains. To …

    manitoba Repository record for Privacy-preserving synthetic image data generation and classification (opens in a new tab)

Page 1 of 3