Copertina

Back to Square Roots: An Optimal Bound on the Matrix Factorization Error for Multi-Epoch Differentially Private SGD

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Kalinin, Nikita P., McKenna, Ryan, Upadhyay, Jalaj, Lampert, Christoph H.
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Cryptography and Security Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Normalized Square Root: Sharper Matrix Factorization Bounds for Differentially Private Continual Counting
di: Henzinger, Monika, et al.
Pubblicazione: (2025)

Banded Square Root Matrix Factorization for Differentially Private Model Training
di: Kalinin, Nikita P., et al.
Pubblicazione: (2024)

Scaling up the Banded Matrix Factorization Mechanism for Differentially Private ML
di: McKenna, Ryan
Pubblicazione: (2024)

Almost Tight Error Bounds on Differentially Private Continual Counting
di: Henzinger, Monika, et al.
Pubblicazione: (2022)

DP-KAN: Differentially Private Kolmogorov-Arnold Networks
di: Kalinin, Nikita P., et al.
Pubblicazione: (2024)

Differentially Private Decentralized Learning with Random Walks
di: Cyffers, Edwige, et al.
Pubblicazione: (2024)

Near-Optimal Generalized Private Testing
di: Chaturvedi, Anamay, et al.
Pubblicazione: (2026)

Benchmarking Differentially Private Tabular Data Synthesis
di: Chen, Kai, et al.
Pubblicazione: (2025)

Optimality of Matrix Mechanism on $\ell_p^p$-metric
di: Liu, Jingcheng, et al.
Pubblicazione: (2024)

Correlated Noise Mechanisms for Differentially Private Learning
di: Pillutla, Krishna, et al.
Pubblicazione: (2025)

ACTG-ARL: Differentially Private Conditional Text Generation with RL-Boosted Control
di: Hu, Yuzheng, et al.
Pubblicazione: (2025)

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

Sampling-Free Privacy Accounting for Matrix Mechanisms under Random Allocation
di: Schuchardt, Jan, et al.
Pubblicazione: (2026)

Is API Access to LLMs Useful for Generating Private Synthetic Tabular Data?
di: Swanberg, Marika, et al.
Pubblicazione: (2025)

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

Scaling Laws for Differentially Private Language Models
di: McKenna, Ryan, et al.
Pubblicazione: (2025)

Differentially Private Federated $k$-Means Clustering with Server-Side Data
di: Scott, Jonathan, et al.
Pubblicazione: (2025)

Beyond Square Roots: Explicit Memory-Efficient Factorization for Multi-Epoch Private Learning
di: Kalinin, Nikita P., et al.
Pubblicazione: (2026)

Rethinking the Security of DP-SGD: A Corrected Analysis of Differentially Private Machine Learning
di: Wang, Wenhao, et al.
Pubblicazione: (2026)

Releasing Large-Scale Human Mobility Histograms with Differential Privacy
di: Bian, Christopher, et al.
Pubblicazione: (2024)

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

On the Price of Differential Privacy for Hierarchical Clustering
di: Deng, Chengyuan, et al.
Pubblicazione: (2025)

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

Differentially Private Substring and Document Counting with Near-Optimal Error
di: Bernardini, Giulia, et al.
Pubblicazione: (2024)

MAPLE: Metadata Augmented Private Language Evolution
di: Chien, Eli, et al.
Pubblicazione: (2026)

Binned Group Algebra Factorization for Differentially Private Continual Counting
di: Henzinger, Monika, et al.
Pubblicazione: (2025)

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

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

Almost linear time differentially private release of synthetic graphs
di: Liu, Jingcheng, et al.
Pubblicazione: (2024)

GeoClip: Geometry-Aware Clipping for Differentially Private SGD
di: Gilani, Atefeh, 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)

Continual Release Moment Estimation with Differential Privacy
di: Kalinin, Nikita P., et al.
Pubblicazione: (2025)

DP-λCGD: Efficient Noise Correlation for Differentially Private Model Training
di: Kalinin, Nikita P., et al.
Pubblicazione: (2026)

VaultGemma: A Differentially Private Gemma Model
di: Sinha, Amer, et al.
Pubblicazione: (2025)

Differentially Private Contrastive Learning via Bounding Group-level Contribution
di: Li, Kecen, et al.
Pubblicazione: (2026)

Interpreting the Error of Differentially Private Median Queries through Randomization Intervals
di: Humphries, Thomas, et al.
Pubblicazione: (2026)

Differentially Private Quasi-Concave Optimization: Bypassing the Lower Bound and Application to Geometric Problems
di: Nissim, Kobbi, et al.
Pubblicazione: (2025)

Differentially Private Multi-objective Selection: Pareto and Aggregation Approaches
di: Farias, Victor A. E., et al.
Pubblicazione: (2024)

Private Count Release: A Simple and Scalable Approach for Private Data Analytics
di: Rogers, Ryan
Pubblicazione: (2024)