Copertina

An Improved Privacy and Utility Analysis of Differentially Private SGD with Bounded Domain and Smooth Losses

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Liang, Hao, Zhang, Wanrong, He, Xinlei, Wu, Kaishun, Xing, Hong
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Machine Learning Cryptography and Security
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Differential Privacy as a Perk: Federated Learning over Multiple-Access Fading Channels with a Multi-Antenna Base Station
di: Liang, Hao, et al.
Pubblicazione: (2025)

PCDP-SGD: Improving the Convergence of Differentially Private SGD via Projection in Advance
di: Sha, Haichao, et al.
Pubblicazione: (2023)

Convergent Privacy Loss of Noisy-SGD without Convexity and Smoothness
di: Chien, Eli, et al.
Pubblicazione: (2024)

Differentially Private SGD Without Clipping Bias: An Error-Feedback Approach
di: Zhang, Xinwei, et al.
Pubblicazione: (2023)

Fundamental Limitations of Favorable Privacy-Utility Guarantees for DP-SGD
di: Ertan, Murat Bilgehan, et al.
Pubblicazione: (2026)

Improving the Utility of Differentially Private Clustering through Dynamical Processing
di: Byun, Junyoung, et al.
Pubblicazione: (2023)

Back to Square Roots: An Optimal Bound on the Matrix Factorization Error for Multi-Epoch Differentially Private SGD
di: Kalinin, Nikita P., et al.
Pubblicazione: (2025)

The Last Iterate Advantage: Empirical Auditing and Principled Heuristic Analysis of Differentially Private SGD
di: Steinke, Thomas, et al.
Pubblicazione: (2024)

DPIS: An Enhanced Mechanism for Differentially Private SGD with Importance Sampling
di: Wei, Jianxin, et al.
Pubblicazione: (2022)

It's Our Loss: No Privacy Amplification for Hidden State DP-SGD With Non-Convex Loss
di: Annamalai, Meenatchi Sundaram Muthu Selva
Pubblicazione: (2024)

PLRV-O: Advancing Differentially Private Deep Learning via Privacy Loss Random Variable Optimization
di: Yang, Qin, et al.
Pubblicazione: (2025)

Improving the Privacy and Practicality of Objective Perturbation for Differentially Private Linear Learners
di: Redberg, Rachel, et al.
Pubblicazione: (2023)

DOPPLER: Differentially Private Optimizers with Low-pass Filter for Privacy Noise Reduction
di: Zhang, Xinwei, et al.
Pubblicazione: (2024)

Augment then Smooth: Reconciling Differential Privacy with Certified Robustness
di: Wu, Jiapeng, et al.
Pubblicazione: (2023)

Private Linear Regression with Differential Privacy and PAC Privacy
di: Yang, Hillary, et al.
Pubblicazione: (2024)

RQP-SGD: Differential Private Machine Learning through Noisy SGD and Randomized Quantization
di: Feng, Ce, et al.
Pubblicazione: (2024)

Safeguarding Data in Multimodal AI: A Differentially Private Approach to CLIP Training
di: Huang, Alyssa, et al.
Pubblicazione: (2023)

DC-SGD: Differentially Private SGD with Dynamic Clipping through Gradient Norm Distribution Estimation
di: Wei, Chengkun, et al.
Pubblicazione: (2025)

GeoClip: Geometry-Aware Clipping for Differentially Private SGD
di: Gilani, Atefeh, et al.
Pubblicazione: (2025)

A Differentially Private Kaplan-Meier Estimator for Privacy-Preserving Survival Analysis
di: Veeraragavan, Narasimha Raghavan, et al.
Pubblicazione: (2024)

Differentially Private Selection using Smooth Sensitivity
di: Chaves, Iago, et al.
Pubblicazione: (2025)

Bounding the Excess Risk for Linear Models Trained on Marginal-Preserving, Differentially-Private, Synthetic Data
di: Zhou, Yvonne, et al.
Pubblicazione: (2024)

Differentially Private Domain Adaptation with Theoretical Guarantees
di: Bassily, Raef, et al.
Pubblicazione: (2023)

Delving into Differentially Private Transformer
di: Ding, Youlong, et al.
Pubblicazione: (2024)

Differentially Private Relational Learning with Entity-level Privacy Guarantees
di: Huang, Yinan, et al.
Pubblicazione: (2025)

Calibrating Practical Privacy Risks for Differentially Private Machine Learning
di: Gu, Yuechun, et al.
Pubblicazione: (2024)

Clients Collaborate: Flexible Differentially Private Federated Learning with Guaranteed Improvement of Utility-Privacy Trade-off
di: Li, Yuecheng, et al.
Pubblicazione: (2024)

SMOTE-DP: Improving Privacy-Utility Tradeoff with Synthetic Data
di: Zhou, Yan, et al.
Pubblicazione: (2025)

Membership Inference Attacks and Privacy in Topic Modeling
di: Manzonelli, Nico, et al.
Pubblicazione: (2024)

Privacy-Preserving In-Context Learning with Differentially Private Few-Shot Generation
di: Tang, Xinyu, et al.
Pubblicazione: (2023)

Differentially Private Active Learning: Balancing Effective Data Selection and Privacy
di: Schwethelm, Kristian, et al.
Pubblicazione: (2024)

On Mitigating the Utility-Loss in Differentially Private Learning: A new Perspective by a Geometrically Inspired Kernel Approach
di: Kumar, Mohit, et al.
Pubblicazione: (2023)

Smooth Lower Bounds for Differentially Private Algorithms via Padding-and-Permuting Fingerprinting Codes
di: Peter, Naty, et al.
Pubblicazione: (2023)

Graphical vs. Deep Generative Models: Measuring the Impact of Differentially Private Mechanisms and Budgets on Utility
di: Ganev, Georgi, et al.
Pubblicazione: (2023)

Privacy Amplification by Structured Subsampling for Deep Differentially Private Time Series Forecasting
di: Schuchardt, Jan, et al.
Pubblicazione: (2025)

Privacy Leakage via Output Label Space and Differentially Private Continual Learning
di: Tobaben, Marlon, et al.
Pubblicazione: (2024)

FlashDP: Private Training Large Language Models with Efficient DP-SGD
di: Wang, Liangyu, et al.
Pubblicazione: (2025)

Less Random, More Private: What is the Optimal Subsampling Scheme for DP-SGD?
di: Dong, Andy, et al.
Pubblicazione: (2026)

How Private are DP-SGD Implementations?
di: Chua, Lynn, et al.
Pubblicazione: (2024)

Can Differentially Private Fine-tuning LLMs Protect Against Privacy Attacks?
di: Du, Hao, et al.
Pubblicazione: (2025)