Cover Image

Gradients Look Alike: Sensitivity is Often Overestimated in DP-SGD

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Thudi, Anvith, Jia, Hengrui, Meehan, Casey, Shumailov, Ilia, Papernot, Nicolas
Format:	Preprint
Published:	2023
Subjects:	Machine Learning Artificial Intelligence Cryptography and Security
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Fast Exact Unlearning for In-Context Learning Data for LLMs
by: Muresanu, Andrei I., et al.
Published: (2024)

Have it your way: Individualized Privacy Assignment for DP-SGD
by: Boenisch, Franziska, et al.
Published: (2023)

Tighter Privacy Auditing of DP-SGD in the Hidden State Threat Model
by: Cebere, Tudor, et al.
Published: (2024)

Architectural Neural Backdoors from First Principles
by: Langford, Harry, et al.
Published: (2024)

Efficient Public Verification of Private ML via Regularization
by: Bell, Zoë Ruha, et al.
Published: (2025)

Backdoor Detection through Replicated Execution of Outsourced Training
by: Jia, Hengrui, et al.
Published: (2025)

Architectural Backdoors for Within-Batch Data Stealing and Model Inference Manipulation
by: Küchler, Nicolas, et al.
Published: (2025)

Machine Learning needs Better Randomness Standards: Randomised Smoothing and PRNG-based attacks
by: Dahiya, Pranav, et al.
Published: (2023)

Beyond Labeling Oracles: What does it mean to steal ML models?
by: Shafran, Avital, et al.
Published: (2023)

When Vision Fails: Text Attacks Against ViT and OCR
by: Boucher, Nicholas, et al.
Published: (2023)

Beyond Laplace and Gaussian: Exploring the Generalized Gaussian Mechanism for Private Machine Learning
by: Rinberg, Roy, et al.
Published: (2025)

Inexact Unlearning Needs More Careful Evaluations to Avoid a False Sense of Privacy
by: Hayes, Jamie, et al.
Published: (2024)

Beyond Slow Signs in High-fidelity Model Extraction
by: Foerster, Hanna, et al.
Published: (2024)

The Curse of Recursion: Training on Generated Data Makes Models Forget
by: Shumailov, Ilia, et al.
Published: (2023)

UnUnlearning: Unlearning is not sufficient for content regulation in advanced generative AI
by: Shumailov, Ilia, et al.
Published: (2024)

Breach By A Thousand Leaks: Unsafe Information Leakage in `Safe' AI Responses
by: Glukhov, David, et al.
Published: (2024)

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

Stealing User Prompts from Mixture of Experts
by: Yona, Itay, et al.
Published: (2024)

Trusted Machine Learning Models Unlock Private Inference for Problems Currently Infeasible with Cryptography
by: Shumailov, Ilia, et al.
Published: (2025)

Reasoning Introduces New Poisoning Attacks Yet Makes Them More Complicated
by: Foerster, Hanna, et al.
Published: (2025)

Interpreting the Repeated Token Phenomenon in Large Language Models
by: Yona, Itay, et al.
Published: (2025)

Locking Machine Learning Models into Hardware
by: Clifford, Eleanor, et al.
Published: (2024)

LLM Dataset Inference: Did you train on my dataset?
by: Maini, Pratyush, et al.
Published: (2024)

Honeyval: A Comprehensive Evaluation Framework for LLM-powered HTTP Honeypots
by: Vero, Mark, et al.
Published: (2026)

The Erasure Illusion: Stress-Testing the Generalization of LLM Forgetting Evaluation
by: Jia, Hengrui, et al.
Published: (2025)

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

DP-SGD with weight clipping
by: Barczewski, Antoine, et al.
Published: (2023)

To Shuffle or not to Shuffle: Auditing DP-SGD with Shuffling
by: Annamalai, Meenatchi Sundaram Muthu Selva, et al.
Published: (2024)

FlashDP: Private Training Large Language Models with Efficient DP-SGD
by: Wang, Liangyu, et al.
Published: (2025)

SEA: Shareable and Explainable Attribution for Query-based Black-box Attacks
by: Gao, Yue, et al.
Published: (2023)

Buffer Overflow in Mixture of Experts
by: Hayes, Jamie, et al.
Published: (2024)

Gaussian DP for Reporting Differential Privacy Guarantees in Machine Learning
by: Gomez, Juan Felipe, et al.
Published: (2025)

DP-TabICL: In-Context Learning with Differentially Private Tabular Data
by: Carey, Alycia N., et al.
Published: (2024)

Too Good to be True? Turn Any Model Differentially Private With DP-Weights
by: Zagardo, David
Published: (2024)

R+R:Understanding Hyperparameter Effects in DP-SGD
by: Morsbach, Felix, et al.
Published: (2024)

Tight Group-Level DP Guarantees for DP-SGD with Sampling via Mixture of Gaussians Mechanisms
by: Ganesh, Arun
Published: (2024)

Confidential Guardian: Cryptographically Prohibiting the Abuse of Model Abstention
by: Rabanser, Stephan, et al.
Published: (2025)

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

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

DP-TRAE: A Dual-Phase Merging Transferable Reversible Adversarial Example for Image Privacy Protection
by: Du, Xia, et al.
Published: (2025)