The paper introduces a simple mechanism that adds a random number of dummy points to protect batch size information in Differentially Private Stochastic Gradient Descent (DP‑SGD). This approach addresses privacy amplification challenges arising from variable batch sizes and truncation in both centralized and distributed settings. The authors provide a general analysis for Poisson sampling and random allocation, showing that the expected number of dummy points is asymptotically smaller than directly privatizing batch sizes, and demonstrate empirically that their method outperforms recent truncation accounting techniques.
arXiv:2602. 17284v2 Announce Type: replace Abstract: We consider the privacy amplification properties of a sampling scheme in which a user's data isused in $k$ steps chosen randomly and uniformly from a sequence (or set) of $t$ steps.
arXiv:2610. 42117v1 Announce Type: new Abstract: In the random allocation sampling scheme each user's data is used in $k$ steps chosen randomly and uniformly from a sequence of $t$ steps.
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.
arXiv:2606. 02563v1 Announce Type: new Abstract: Heterogeneous Differential Privacy (HDP) in Federated Learning (FL) allows clients to select individual privacy budgets ($\varepsilon_i$) according to institutional policies and data sensitivity.
Heterogeneous Differential Privacy (HDP) in Federated Learning (FL) allows clients to select individual privacy budgets ($\varepsilon_i$) according to institutional policies and data sensitivity. In practice, many HDP-FL systems employ $\varepsilon$-aware server aggregation to improve model utility by re-weighting client updates according to their declared privacy budgets.
arXiv:2605. 11170v2 Announce Type: replace Abstract: Noise-based certified machine unlearning currently faces a hard ceiling: the noise magnitude required to certify unlearning typically destroys model utility, particularly for large-scale deletion requests.
arXiv:2510. 04902v3 Announce Type: replace Abstract: Tuning hyperparameters in federated machine learning can substantially impact model performance.
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.
arXiv:2411.16478v3 Announce Type: replace Abstract: Measuring distribution drifts is a key task in managing distributed, sensitive data, as it underpins a wide range of federated learning and analyti...
arXiv:2608. 15153v1 Announce Type: cross Abstract: Differentially private federated learning must balance privacy protection against model accuracy and training efficiency.
arXiv:2505. 22703v2 Announce Type: replace Abstract: Many problems in trustworthy ML can be expressed as constraints on prediction rates across subpopulations, including group fairness constraints (demographic parity, equalized odds, etc.
arXiv:2610.27710v1 Announce Type: new Abstract: Diffusion models have emerged as the foundation of modern generative systems, yet their high memorization capacity raises privacy concerns. While diffe...