arXiv Machine Learning October 5, 2026 By Marten van Dijk, Murat Bilgehan Ertan

Trade-off Functions for DP-SGD with Subsampling based on Random Allocation: Tight Upper and Lower Bounds

Read the original on arXiv Machine Learning →

arXiv:2605. 06259v3 Announce Type: replace Abstract: Within the $f$-DP framework, we derive a tight analysis of the trade-off function for Differentially Private Stochastic Gradient Descent (DP-SGD) with subsampling based on random allocation in which each sample is independently assigned to exactly one of $M$ minibatches per epoch, each minibatch corresponding to one of the $M$ SGD rounds within a single epoch.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

safety

arXiv Machine Learning

Aug 11

Fundamental Limitations of Favorable Privacy-Utility Guarantees for DP-SGD

arXiv:2601. 10237v3 Announce Type: replace Abstract: Differentially Private Stochastic Gradient Descent (DP-SGD) is the dominant paradigm for private training, but its fundamental limitations under worst-case adversarial privacy definitions remain poorly understood.

By Murat Bilgehan Ertan, Marten van Dijk

safety

arXiv Machine Learning

Sep 18

Near-Optimal Machine Unlearning Utility for Smooth Strongly Convex Losses

arXiv:2606. 01527v2 Announce Type: replace Abstract: Machine unlearning is motivated by legal and user-facing requirements to remove the influence of individuals' data from trained models, such as the right to be forgotten.

By Matthew Regehr, Gautam Kamath, Andrew Lowy

safety

arXiv Machine Learning

1d ago

Closed-Form Noise Calibration Against Membership Inference for Random-Allocation DP-SGD

arXiv:2610. 09651v1 Announce Type: new Abstract: DP-SGD protects training data by adding Gaussian noise to clipped gradients.

By Murat Bilgehan Ertan, Marten van Dijk

benchmarks safety

arXiv Machine Learning

Jun 5

Minimax optimal differentially private synthetic data for smooth queries

arXiv:2602. 01607v3 Announce Type: replace-cross Abstract: Differentially private synthetic data enables the sharing and analysis of sensitive datasets while providing rigorous privacy guarantees for individual contributors.

By Rundong Ding, Yiyun He, Yizhe Zhu

safety

arXiv Machine Learning

Sep 22

Improved Private Sparse Covariance Estimation with Multiscale Threshold Tests

arXiv:2609. 22783v1 Announce Type: new Abstract: We study differentially private covariance estimation in operator norm for mean-zero sub-Gaussian distributions with unknown covariance support and at most $k$ nonzero entries per row.

By Zihan Zhang

safety

arXiv Machine Learning

Jul 10

On the Gradient Complexity of Private Optimization with Private Oracles

arXiv:2511. 13999v2 Announce Type: replace Abstract: We study the running time, in terms of first order oracle queries, of differentially private empirical/population risk minimization of Lipschitz convex losses.

By Michael Menart, Aleksandar Nikolov

efficiency

One email a morning, machine-written

Related stories

Fundamental Limitations of Favorable Privacy-Utility Guarantees for DP-SGD

Near-Optimal Machine Unlearning Utility for Smooth Strongly Convex Losses

Closed-Form Noise Calibration Against Membership Inference for Random-Allocation DP-SGD

Minimax optimal differentially private synthetic data for smooth queries

Improved Private Sparse Covariance Estimation with Multiscale Threshold Tests

On the Gradient Complexity of Private Optimization with Private Oracles