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 66 for “"Differential Privacy (DP)"”.

  1. Adversarial Attacks on Differentially Private Factorization Machines in Recommendation Systems

    … large amounts of user data, raising significant privacy concerns. Differential Privacy (DP) has emerged as a viable solution that provides privacy guarantees by introducing controlled noise into the model computations. Despite these protections, differentially private FM models remain vulnerable …

    ttu Repository record for Adversarial Attacks on Differentially Private Factorization Machines in Recommendation Systems (opens in a new tab)

  2. ALGORITHMIC IMPROVEMENTS FOR DIFFERENTIALLY PRIVATE LEARNING

    … information when learning from private data. Differential privacy (DP) provides rigorous protection for learning algorithms, yet applying DP to neural network learning remains challenging due to diverse data types and complex architectures. To address these challenges, this thesis introduces …

    nus Repository record for ALGORITHMIC IMPROVEMENTS FOR DIFFERENTIALLY PRIVATE LEARNING (opens in a new tab) Download file for ALGORITHMIC IMPROVEMENTS FOR DIFFERENTIALLY PRIVATE LEARNING (opens in a new tab)

  3. EFFICIENT AND PRIVACY-PRESERVING FEDERATED LEARNING FOR RECOMMENDER SYSTEMS

    … centralized data collection, raising significant privacy concerns. In light of these concerns, federated recommender systems (FedRec) emerges as a potential solution that enables collaborative model training without sharing raw user data. However, existing FedRec approaches face challenges in …

    nus Repository record for EFFICIENT AND PRIVACY-PRESERVING FEDERATED LEARNING FOR RECOMMENDER SYSTEMS (opens in a new tab) Download file for EFFICIENT AND PRIVACY-PRESERVING FEDERATED LEARNING FOR RECOMMENDER SYSTEMS (opens in a new tab)

  4. Privacy-Preserving Edge-Cloud Architecture for IoT Healthcare Systems

    … (IoT) healthcare applications, a myriad of data privacy concerns come to light. Cloud computing inherits the risks of exposing data to re-identification vulnerabilities. A secure solution is storing and processing data locally on edge, but it lacks the provision of powerful machine learning (ML) …

    york Repository record for Privacy-Preserving Edge-Cloud Architecture for IoT Healthcare Systems (opens in a new tab)

  5. DProvSQL: Accuracy-Aware Privacy Provenance Framework for Differentially Private SQL Engine

    Recent years have witnessed the adoption of differential privacy (DP) in practical database query systems. Such systems, like PrivateSQL and FLEX, allow data analysts to query sensitive data while providing a rigorous and provable privacy guarantee. However, existing systems may use more privacy …

    waterloo Repository record for DProvSQL: Accuracy-Aware Privacy Provenance Framework for Differentially Private SQL Engine (opens in a new tab)

  6. PRIVATE DENSITY ESTIMATION FOR MIXTURE DISTRIBUTIONS AND GAUSSIAN MIXTURE MODELS

    … (mixture) distributions under the constraint of differential privacy (DP). On a high level, we show that if a class of distributions (such as Gaussians) is (1) list decodable and (2) admits a “locally small” cover (Bun et al., 2021) with respect to total variation distance, then the class of its …

    mcmaster Repository record for PRIVATE DENSITY ESTIMATION FOR MIXTURE DISTRIBUTIONS AND GAUSSIAN MIXTURE MODELS (opens in a new tab)

  7. A kernel-based approach to differentially private image generation

    The gold standard privacy notion, differential privacy (DP), has gained widespread adoption in academic research, industry products, and government databases due to its mathematically provable privacy guarantee. However, the composability property of DP leads to privacy degradation with multiple …

    ubc Repository record for A kernel-based approach to differentially private image generation (opens in a new tab) Download PDF for A kernel-based approach to differentially private image generation (opens in a new tab)

  8. Revisiting Benchmarks for Privacy-Preserving Image Classification

    Differential privacy (DP) is a standard method for preserving the privacy of individual data points. DP prevents models from memorizing training data, thus reducing the risk of data leakage. While DP has been effective in machine learning (ML), there are growing concerns about some common practices …

    waterloo Repository record for Revisiting Benchmarks for Privacy-Preserving Image Classification (opens in a new tab)

  9. Differentially Private Approximation Algorithms for Facility Location Problem

    … the constraints imposed by central and local differential privacy (LDP) models. Recently, proposed lower and upper bounds for the problem on the central differential privacy (DP) model problem, where a trusted curator first collects and processes all data. This thesis focuses on the DP and the …

    buffalo Repository record for Differentially Private Approximation Algorithms for Facility Location Problem (opens in a new tab)

  10. Differentially-private Multiparty Clustering

    … (ML) across diverse domains, the necessity of privacy-preserving techniques has become paramount. The Euclidean k-Means problem, a fundamental component of unsupervised learning, brings to light this privacy challenge, especially in federated contexts. Existing Federated approaches utilizing …

    waterloo Repository record for Differentially-private Multiparty Clustering (opens in a new tab)

  11. Sensitivity computation for user-defi ned functions in Differential Privacy systems

    Differential privacy (DP) is emerging as a viable solution to release statistical information about a population without compromising data subjects’ privacy. A standard way to achieve DP is by adding calibrated noise to the result of some statistical analysis. To account for the impact an …

    chalmers Repository record for Sensitivity computation for user-defi ned functions in Differential Privacy systems (opens in a new tab)

  12. Practical Privacy-Preserving Federated Learning with Secure Multi-Party Computation

    … is still susceptible to attacks that compromise privacy through privacy inference and poisoning attacks. Through secure Multi-Party Computation (MPC) and Differential Privacy (DP) it is possible to eliminate more of these attacks. Secure Multi-Party Computation is a promising efficient solution …

    vt Repository record for Practical Privacy-Preserving Federated Learning with Secure Multi-Party Computation (opens in a new tab)

  13. Selective Metric Differential Privacy for Language Models

    … the large text corpora they were trained on. In privacy-preserving machine learning, it is common to apply Differential Privacy (DP) mechanisms that mitigate such leakage. The traditional notion of DP, where each record in the data is treated as sensitive, does not translate well to NLP tasks …

    washington Repository record for Selective Metric Differential Privacy for Language Models (opens in a new tab)

  14. Ancestry Deconvolution via Differential Privacy

    … the first study of ancestry determination under differential privacy (DP). Direct-to-consumer genomics companies, such as 23andMe, offer ancestry testing to millions of individuals, yet remain vulnerable to severe data breaches. Such incidents are especially concerning because genomic data is …

    waterloo Repository record for Ancestry Deconvolution via Differential Privacy (opens in a new tab)

  15. SEDPD: Sampling-Enhanced Differentially Private Defense against Backdoor Poisoning Attacks of Image Classification

    … as SHAP and LIME. These perturbations ensure differential privacy(DP) bounds, disrupting adversarial attempts to embed malicious triggers while maintaining quality of explanation-values for legitimate users. To validate our defense, we introduce a threat model tailored to image classification …

    ku Repository record for SEDPD: Sampling-Enhanced Differentially Private Defense against Backdoor Poisoning Attacks of Image Classification (opens in a new tab)

  16. Privacy meets Robustness: Unveiling the interplay between Differential Privacy and Robustness in Machine Learning

    … has raised critical concerns regarding the privacy of individuals whose data is being used, as well as the robustness of algorithms against potentially malicious data corruption from unreliable sources. This thesis aims to explore the fundamental interplay between differential privacy (DP) …

    washington Repository record for Privacy meets Robustness: Unveiling the interplay between Differential Privacy and Robustness in Machine Learning (opens in a new tab)

  17. Analysis of Privacy-aware Data Sharing in Cyber-physical Energy Systems

    … the key factors and correlations among the privacy, security, and utility requirements of grid networks to ensure effective inter-and intra-actions within physical layer equipment (e.g., distributed energy resources (DERs), intelligent electronic devices (IEDs), etc.). We have conducted a …

    unr Repository record for Analysis of Privacy-aware Data Sharing in Cyber-physical Energy Systems (opens in a new tab)

  18. Application Of Blockchain Technology And Integration Of Differential Privacy: Issues In E-Health Domains

    … applications of Blockchain Technology with Differential Privacy integration lies within privacy and security enhancement. This paper aims to highlight the research issues in the e-Health domain (e.g., EMR) and to review the current research directions in Differential Privacy integration with …

    nodak Repository record for Application Of Blockchain Technology And Integration Of Differential Privacy: Issues In E-Health Domains (opens in a new tab)

  19. Improving Privacy-utility Tradeoffs in Privacy-preserving Data Release with Context Information

    … pervasive, there's a growing need to incorporate privacy into the data consumption pipeline. Furthermore, as data acquisition becomes increasingly personalized and tailored to specific applications, contextual knowledge about the underlying data often becomes available, such as prior …

    arizona-thes Repository record for Improving Privacy-utility Tradeoffs in Privacy-preserving Data Release with Context Information (opens in a new tab)

  20. Adaptive Differential Privacy Budgeting Strategy for Optimizing Synthetic Data Generation and Privacy–Utility Trade-offs

    Training generative models under differential privacy (DP) requires injecting calibrated noise into gradient updates, creating an inherent trade-off between privacy protection and data quality. In standard DP-CTGAN, a single discriminator processes all features under a shared privacy budget, so …

    waterloo Repository record for Adaptive Differential Privacy Budgeting Strategy for Optimizing Synthetic Data Generation and Privacy–Utility Trade-offs (opens in a new tab)

Page 1 of 4