Cover Image

Clip Body and Tail Separately: High Probability Guarantees for DPSGD with Heavy Tails

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Sha, Haichao, Cao, Yang, Liu, Yong, Wu, Yuncheng, Liu, Ruixuan, Chen, Hong
Format:	Preprint
Published:	2024
Subjects:	Machine Learning Cryptography and Security
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

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

Privacy of SGD under Gaussian or Heavy-Tailed Noise: Guarantees without Gradient Clipping
by: Şimşekli, Umut, et al.
Published: (2024)

DP-SelFT: Differentially Private Selective Fine-Tuning for Large Language Models
by: Sha, Haichao, et al.
Published: (2026)

Weights Shuffling for Improving DPSGD in Transformer-based Models
by: Yang, Jungang, et al.
Published: (2024)

Optimal Rates for Pure $\varepsilon$-Differentially Private Stochastic Convex Optimization with Heavy Tails
by: Lowy, Andrew
Published: (2026)

Private Stochastic Convex Optimization with Heavy Tails: Near-Optimality from Simple Reductions
by: Asi, Hilal, et al.
Published: (2024)

FERRET: Private Deep Learning Faster And Better Than DPSGD
by: Zagardo, David
Published: (2025)

Enhancing DPSGD via Per-Sample Momentum and Low-Pass Filtering
by: Xu, Xincheng, et al.
Published: (2025)

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

DPDR: Gradient Decomposition and Reconstruction for Differentially Private Deep Learning
by: Liu, Yixuan, et al.
Published: (2024)

The Tail Tells All: Estimating Model-Level Membership Inference Vulnerability Without Reference Models
by: Dodd, Euodia, et al.
Published: (2025)

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

DP-SGD Without Clipping: The Lipschitz Neural Network Way
by: Bethune, Louis, et al.
Published: (2023)

Clipped SGD Algorithms for Performative Prediction: Tight Bounds for Clipping Bias and Remedies
by: Li, Qiang, et al.
Published: (2024)

Passive Inference Attacks on Split Learning via Adversarial Regularization
by: Zhu, Xiaochen, et al.
Published: (2023)

Beyond Indistinguishability: Measuring Extraction Risk in LLM APIs
by: Liu, Ruixuan, et al.
Published: (2026)

Enabling Differentially Private Federated Learning for Speech Recognition: Benchmarks, Adaptive Optimizers and Gradient Clipping
by: Pelikan, Martin, et al.
Published: (2023)

Efficient Unlearning with Privacy Guarantees
by: Domingo-Ferrer, Josep, et al.
Published: (2025)

On the Convergence of DP-SGD with Adaptive Clipping
by: Shulgin, Egor, et al.
Published: (2024)

Membership Inference Attacks on Large-Scale Models: A Survey
by: Wu, Hengyu, et al.
Published: (2025)

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

Clip-and-Verify: Linear Constraint-Driven Domain Clipping for Accelerating Neural Network Verification
by: Zhou, Duo, et al.
Published: (2025)

Log Probability Tracking of LLM APIs
by: Chauvin, Timothée, et al.
Published: (2025)

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

Auditing Differential Privacy Guarantees Using Density Estimation
by: Koskela, Antti, et al.
Published: (2024)

A New Analysis of Differential Privacy's Generalization Guarantees
by: Jung, Christopher, et al.
Published: (2019)

FedGIG: Graph Inversion from Gradient in Federated Learning
by: Xiao, Tianzhe, et al.
Published: (2024)

Long-Tailed Backdoor Attack Using Dynamic Data Augmentation Operations
by: Pang, Lu, et al.
Published: (2024)

Revisiting ML Training under Fully Homomorphic Encryption: Convergence Guarantees, Differential Privacy, and Efficient Algorithms
by: Zhou, Yvonne, et al.
Published: (2026)

Detecting Credit Card Fraud via Heterogeneous Graph Neural Networks with Graph Attention
by: Sha, Qiuwu, et al.
Published: (2025)

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

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

DP-Adam-AC: Privacy-preserving Fine-Tuning of Localizable Language Models Using Adam Optimization with Adaptive Clipping
by: Yang, Ruoxing
Published: (2025)

FedRE: Robust and Effective Federated Learning with Privacy Preference
by: Xiao, Tianzhe, et al.
Published: (2025)

Towards Certified Malware Detection: Provable Guarantees Against Evasion Attacks
by: Giri, Nandakrishna, et al.
Published: (2026)

Federated Online Prediction from Experts with Differential Privacy: Separations and Regret Speed-ups
by: Gao, Fengyu, et al.
Published: (2024)

CBD: A Certified Backdoor Detector Based on Local Dominant Probability
by: Xiang, Zhen, et al.
Published: (2023)

Black-box Optimization of LLM Outputs by Asking for Directions
by: Zhang, Jie, et al.
Published: (2025)

Privacy in Metalearning and Multitask Learning: Modeling and Separations
by: Aliakbarpour, Maryam, et al.
Published: (2024)

DPM: Clustering Sensitive Data through Separation
by: Liebenow, Johannes, et al.
Published: (2023)