Imagen de Portada

The Last Iterate Advantage: Empirical Auditing and Principled Heuristic Analysis of Differentially Private SGD

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Steinke, Thomas, Nasr, Milad, Ganesh, Arun, Balle, Borja, Choquette-Choo, Christopher A., Jagielski, Matthew, Hayes, Jamie, Thakurta, Abhradeep Guha, Smith, Adam, Terzis, Andreas
Formato:	Preprint
Publicado:	2024
Materias:	Cryptography and Security Machine Learning
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Privacy Amplification for Matrix Mechanisms
por: Choquette-Choo, Christopher A., et al.
Publicado: (2023)

Auditing Private Prediction
por: Chadha, Karan, et al.
Publicado: (2024)

Optimal Rates for $O(1)$-Smooth DP-SCO with a Single Epoch and Large Batches
por: Choquette-Choo, Christopher A., et al.
Publicado: (2024)

Correlated Noise Provably Beats Independent Noise for Differentially Private Learning
por: Choquette-Choo, Christopher A., et al.
Publicado: (2023)

Near Exact Privacy Amplification for Matrix Mechanisms
por: Choquette-Choo, Christopher A., et al.
Publicado: (2024)

On Design Principles for Private Adaptive Optimizers
por: Ganesh, Arun, et al.
Publicado: (2025)

Hush! Protecting Secrets During Model Training: An Indistinguishability Approach
por: Ganesh, Arun, et al.
Publicado: (2025)

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

Correlated Noise Mechanisms for Differentially Private Learning
por: Pillutla, Krishna, et al.
Publicado: (2025)

Privacy Auditing of Large Language Models
por: Panda, Ashwinee, et al.
Publicado: (2025)

InvisibleInk: High-Utility and Low-Cost Text Generation with Differential Privacy
por: Vinod, Vishnu, et al.
Publicado: (2025)

Differentially Private Parameter-Efficient Fine-tuning for Large ASR Models
por: Liu, Hongbin, et al.
Publicado: (2024)

Preserving Expert-Level Privacy in Offline Reinforcement Learning
por: Sharma, Navodita, et al.
Publicado: (2024)

Recycling Scraps: Improving Private Learning by Leveraging Intermediate Checkpoints
por: Shejwalkar, Virat, et al.
Publicado: (2022)

Cascading Adversarial Bias from Injection to Distillation in Language Models
por: Chaudhari, Harsh, et al.
Publicado: (2025)

Privacy Side Channels in Machine Learning Systems
por: Debenedetti, Edoardo, et al.
Publicado: (2023)

LLMs unlock new paths to monetizing exploits
por: Carlini, Nicholas, et al.
Publicado: (2025)

The Attacker Moves Second: Stronger Adaptive Attacks Bypass Defenses Against Llm Jailbreaks and Prompt Injections
por: Nasr, Milad, et al.
Publicado: (2025)

The Hitchhiker's Guide to Efficient, End-to-End, and Tight DP Auditing
por: Annamalai, Meenatchi Sundaram Muthu Selva, et al.
Publicado: (2025)

Differentially Private Stream Processing at Scale
por: Zhang, Bing, et al.
Publicado: (2023)

Reinforced Generation of Combinatorial Structures: Ramsey Numbers
por: Nagda, Ansh, et al.
Publicado: (2026)

Reinforced Generation of Combinatorial Structures: Hardness of Approximation
por: Nagda, Ansh, et al.
Publicado: (2025)

Training Large ASR Encoders with Differential Privacy
por: Chauhan, Geeticka, et al.
Publicado: (2024)

Poisoning Web-Scale Training Datasets is Practical
por: Carlini, Nicholas, et al.
Publicado: (2023)

Efficient and Near-Optimal Noise Generation for Streaming Differential Privacy
por: Dvijotham, Krishnamurthy, et al.
Publicado: (2024)

Phantom: General Backdoor Attacks on Retrieval Augmented Language Generation
por: Chaudhari, Harsh, et al.
Publicado: (2024)

Thought-Transfer: Indirect Targeted Poisoning Attacks on Chain-of-Thought Reasoning Models
por: Chaudhari, Harsh, et al.
Publicado: (2026)

Measuring memorization in language models via probabilistic extraction
por: Hayes, Jamie, et al.
Publicado: (2024)

Beyond the Calibration Point: Mechanism Comparison in Differential Privacy
por: Kaissis, Georgios, et al.
Publicado: (2024)

Synthetic Query Generation for Privacy-Preserving Deep Retrieval Systems using Differentially Private Language Models
por: Carranza, Aldo Gael, et al.
Publicado: (2023)

Lessons from Defending Gemini Against Indirect Prompt Injections
por: Shi, Chongyang, et al.
Publicado: (2025)

Private Gradient Descent for Linear Regression: Tighter Error Bounds and Instance-Specific Uncertainty Estimation
por: Brown, Gavin, et al.
Publicado: (2024)

Extracting alignment data in open models
por: Barbero, Federico, et al.
Publicado: (2025)

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

Scaling Laws for Differentially Private Language Models
por: McKenna, Ryan, et al.
Publicado: (2025)

Are aligned neural networks adversarially aligned?
por: Carlini, Nicholas, et al.
Publicado: (2023)

Differentially private survey research
por: Georgina Evans, et al.
Publicado: (2024)

Portfolio Choice With Intra‐Household Bargaining and Gender Differences in Preferences
por: Amelia Guha Thakurta, et al.
Publicado: (2026)

Fast Last-Iterate Convergence of SGD in the Smooth Interpolation Regime
por: Attia, Amit, et al.
Publicado: (2025)

Last-Iterate Complexity of SGD for Convex and Smooth Stochastic Problems
por: Garrigos, Guillaume, et al.
Publicado: (2025)