Image de couverture de livre

Beyond Square Roots: Explicit Memory-Efficient Factorization for Multi-Epoch Private Learning

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Kalinin, Nikita P., Rehn, Aki, Andersson, Joel Daniel, Honkela, Antti, Lampert, Christoph H.
Format:	Preprint
Publié:	2026
Sujets:	Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

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

Banded Square Root Matrix Factorization for Differentially Private Model Training
par: Kalinin, Nikita P., et autres
Publié: (2024)

On Optimal Hyperparameters for Differentially Private Deep Transfer Learning
par: Rehn, Aki, et autres
Publié: (2025)

Learning Rate Scheduling with Matrix Factorization for Private Training
par: Kalinin, Nikita P., et autres
Publié: (2025)

Mitigating Disparate Impact of Differentially Private Learning through Bounded Adaptive Clipping
par: Zhao, Linzh, et autres
Publié: (2025)

An Interactive Framework for Finding the Optimal Trade-off in Differential Privacy
par: Yang, Yaohong, et autres
Publié: (2025)

Normalized Square Root: Sharper Matrix Factorization Bounds for Differentially Private Continual Counting
par: Henzinger, Monika, et autres
Publié: (2025)

DP-λCGD: Efficient Noise Correlation for Differentially Private Model Training
par: Kalinin, Nikita P., et autres
Publié: (2026)

DP-KAN: Differentially Private Kolmogorov-Arnold Networks
par: Kalinin, Nikita P., et autres
Publié: (2024)

Matrix Factorization for Practical Continual Mean Estimation Under User-Level Differential Privacy
par: Kalinin, Nikita P., et autres
Publié: (2026)

Noise-Aware Differentially Private Variational Inference
par: Alrawajfeh, Talal, et autres
Publié: (2024)

Continual Release Moment Estimation with Differential Privacy
par: Kalinin, Nikita P., et autres
Publié: (2025)

Subsampling is not Magic: Why Large Batch Sizes Work for Differentially Private Stochastic Optimisation
par: Räisä, Ossi, et autres
Publié: (2024)

Efficient and Scalable Implementation of Differentially Private Deep Learning without Shortcuts
par: Beltran, Sebastian Rodriguez, et autres
Publié: (2024)

Binned Group Algebra Factorization for Differentially Private Continual Counting
par: Henzinger, Monika, et autres
Publié: (2025)

A Bias-Variance Decomposition for Ensembles over Multiple Synthetic Datasets
par: Räisä, Ossi, et autres
Publié: (2024)

Privacy Leakage via Output Label Space and Differentially Private Continual Learning
par: Tobaben, Marlon, et autres
Publié: (2024)

DP-MicroAdam: Private and Frugal Algorithm for Training and Fine-tuning
par: Hudişteanu, Mihaela, et autres
Publié: (2025)

Beyond Membership: Limitations of Add/Remove Adjacency in Differential Privacy
par: Pradhan, Gauri, et autres
Publié: (2025)

On Reliability of Efficient Membership Inference Vulnerability Evaluation
par: Jälkö, Joonas, et autres
Publié: (2026)

Noise-Aware Differentially Private Regression via Meta-Learning
par: Räisä, Ossi, et autres
Publié: (2024)

A Smooth Binary Mechanism for Efficient Private Continual Observation
par: Andersson, Joel Daniel, et autres
Publié: (2023)

Notes on Sampled Gaussian Mechanism
par: Kalinin, Nikita P.
Publié: (2024)

Empirical Comparison of Membership Inference Attacks in Deep Transfer Learning
par: Bai, Yuxuan, et autres
Publié: (2025)

Streaming Private Continual Counting via Binning
par: Andersson, Joel Daniel, et autres
Publié: (2024)

Improved Accuracy for Private Continual Cardinality Estimation in Fully Dynamic Streams via Matrix Factorization
par: Andersson, Joel Daniel, et autres
Publié: (2026)

Hyperparameters in Score-Based Membership Inference Attacks
par: Pradhan, Gauri, et autres
Publié: (2025)

Impact of Dataset Properties on Membership Inference Vulnerability of Deep Transfer Learning
par: Tobaben, Marlon, et autres
Publié: (2024)

Population Risk Bounds for Kolmogorov-Arnold Networks Trained by DP-SGD with Correlated Noise
par: Wang, Puyu, et autres
Publié: (2026)

Fast Rate Bounds for Multi-Task and Meta-Learning with Different Sample Sizes
par: Zakerinia, Hossein, et autres
Publié: (2025)

Adaptive Sampling and Clipping for Private Worst-Case Group Optimization
par: Cairney-Leeming, Max, et autres
Publié: (2026)

Sampling-Free Privacy Accounting for Matrix Mechanisms under Random Allocation
par: Schuchardt, Jan, et autres
Publié: (2026)

Differentially Private Federated $k$-Means Clustering with Server-Side Data
par: Scott, Jonathan, et autres
Publié: (2025)

SMMF: Square-Matricized Momentum Factorization for Memory-Efficient Optimization
par: Park, Kwangryeol, et autres
Publié: (2024)

Accuracy-First Rényi Differential Privacy and Post-Processing Immunity
par: Räisä, Ossi, et autres
Publié: (2025)

From Low Intrinsic Dimensionality to Non-Vacuous Generalization Bounds in Deep Multi-Task Learning
par: Zakerinia, Hossein, et autres
Publié: (2025)

Gaussian DP for Reporting Differential Privacy Guarantees in Machine Learning
par: Gomez, Juan Felipe, et autres
Publié: (2025)

Learning Quantized Continuous Controllers for Integer Hardware
par: Kresse, Fabian, et autres
Publié: (2025)

PeFLL: Personalized Federated Learning by Learning to Learn
par: Scott, Jonathan, et autres
Publié: (2023)

More Flexible PAC-Bayesian Meta-Learning by Learning Learning Algorithms
par: Zakerinia, Hossein, et autres
Publié: (2024)