Differential Privacy Guarantee

Differential privacy guarantees aim to mathematically bound the risk of individual data leakage during data analysis or machine learning model training, enabling the release of aggregate results while protecting sensitive information. Current research focuses on improving the privacy-utility trade-off within various settings, including federated learning, synthetic data generation, and deep learning, often employing techniques like noise addition, gradient clipping, and privacy amplification through mechanisms such as shuffling and subsampling. These advancements are crucial for responsible data sharing and the development of privacy-preserving machine learning models across diverse applications, impacting fields ranging from healthcare to finance.

Papers

October 15, 2024

Data-adaptive Differentially Private Prompt Synthesis for In-Context Learning
Fengyu Gao, Ruida Zhou, Tianhao Wang, Cong Shen, Jing Yang
LeArning Abstract Context Learning Different Context Differential Privacy Guarantee Private Datasets Non Private

August 9, 2024

PriPHiT: Privacy-Preserving Hierarchical Training of Deep Neural Networks
Yamin Sepehri, Pedram Pad, Pascal Frossard, L. Andrea Dunbar
Deep Neural Network Privacy Preservation Differential Privacy Guarantee Reconstruction Attack Early Exit Neural Network Hierarchical Training

August 8, 2024

Federated Cubic Regularized Newton Learning with Sparsification-amplified Differential Privacy
Wei Huo, Changxin Liu, Kemi Ding, Karl Henrik Johansson, Ling Shi
Private Federated Learning Differential Privacy Guarantee Federated Learning Framework Newton Method

July 16, 2024

Private prediction for large-scale synthetic text generation
Kareem Amin, Alex Bie, Weiwei Kong, Alexey Kurakin, Natalia Ponomareva, Umar Syed, Andreas Terzis, Sergei Vassilvitskii
Large Language Model Differential Privacy Guarantee Private Prediction Private INFORMATION RETRIEVAL

July 9, 2024

Synthetic Data: Revisiting the Privacy-Utility Trade-off
Fatima Jahan Sarmin, Atiquer Rahman Sarkar, Yang Wang, Noman Mohammed
Synthetic Data Privacy Preserving Privacy Utility Differential Privacy Guarantee Privacy Utility Trade

June 7, 2024

Black Box Differential Privacy Auditing Using Total Variation Distance
Antti Koskela, Jafar Mohammadi
Differential Privacy Black Box Differential Privacy Guarantee Privacy Model Total Variation Distance ONE Shot

May 6, 2024

Differentially Private Federated Learning without Noise Addition: When is it Possible?
Jiang Zhang, Konstantinos Psounis, Salman Avestimehr
Secure Aggregation Private Federated Learning Differential Privacy Guarantee

February 8, 2024

Differentially Private Deep Model-Based Reinforcement Learning
Alexandre Rio, Merwan Barlier, Igor Colin, Albert Thomas
Offline Reinforcement Learning Differential Privacy Guarantee Private Model RL Agent Model Based Policy Private Reinforcement Learning

October 27, 2023

DP-SGD with weight clipping
Antoine Barczewski, Jan Ramon
Data Privacy Differential Privacy Guarantee Weight Update Private Gradient Gradient Clipping

October 24, 2023

Privacy Amplification for Matrix Mechanisms
Christopher A. Choquette-Choo, Arun Ganesh, Thomas Steinke, Abhradeep Thakurta
Privacy Utility Differential Privacy Guarantee Differential Privacy SGD Privacy Amplification

October 13, 2023

Near-optimal Differentially Private Client Selection in Federated Settings
Syed Eqbal Alam, Dhirendra Shukla, Shrisha Rao
Private Algorithm Federated Setting Differential Privacy Guarantee Information Transfer Local Computation Client Level Differential Privacy

September 11, 2023

Share Your Representation Only: Guaranteed Improvement of the Privacy-Utility Tradeoff in Federated Learning
Zebang Shen, Jiayuan Ye, Anmin Kang, Hamed Hassani, Reza Shokri
Individual Representation Large Relevance Improvement Private Data Private Algorithm Privacy Utility Differential Privacy Guarantee Randomized Mechanism Representation Function

September 6, 2023

Roulette: A Semantic Privacy-Preserving Device-Edge Collaborative Inference Framework for Deep Learning Classification Tasks
Jingyi Li, Guocheng Liao, Lin Chen, Xu Chen
Ground Truth Differential Privacy Guarantee Collaborative Inference Deep Learning Task Privacy Preserving Inference

September 1, 2023

Advancing Personalized Federated Learning: Group Privacy, Fairness, and Beyond
Filippo Galli, Kangsoo Jung, Sayan Biswas, Catuscia Palamidessi, Tommaso Cucinotta
Differential Privacy Procedural Fairness Privacy Preserving Machine Learning Differential Privacy Guarantee Group Privacy

August 7, 2023

Binary Federated Learning with Client-Level Differential Privacy
Lumin Liu, Jun Zhang, Shenghui Song, Khaled B. Letaief
Differential Privacy Privacy Preserving Differential Privacy Guarantee Communication Efficient Federated Learning Client Level Differential Privacy

July 20, 2023

Data Analytics with Differential Privacy
Vassilis Digalakis
Differential Privacy Private Algorithm Differential Privacy Guarantee Data Analytics Entropy Estimation

July 11, 2023

Differentially Private Statistical Inference through $\beta$-Divergence One Posterior Sampling
Jack Jewson, Sahra Ghalebikesabi, Chris Holmes
Posterior Sampling Differential Privacy Guarantee Private Estimation Beta Divergence Private Bayesian

June 24, 2023

June 13, 2023

Privacy Preserving Bayesian Federated Learning in Heterogeneous Settings
Disha Makhija, Joydeep Ghosh, Nhat Ho
Privacy Policy Differential Privacy Guarantee Heterogeneous Setting Privacy Constraint Bayesian Federated Learning