Private Federated Learning

Private federated learning aims to enable collaborative model training across multiple devices without compromising data privacy, primarily focusing on mitigating information leakage through model updates. Current research emphasizes techniques like differential privacy (DP), often combined with secure aggregation and various optimization methods (e.g., sharpness-aware minimization, cubic-regularized Newton methods) to enhance the privacy-utility trade-off, and explores efficient compression and quantization strategies to reduce communication overhead. This field is crucial for advancing trustworthy machine learning applications, particularly in sensitive domains like healthcare and finance, by providing rigorous privacy guarantees while maintaining model accuracy.

Papers

November 29, 2022

Adap DP-FL: Differentially Private Federated Learning with Adaptive Noise
Jie Fu, Zhili Chen, Xiao Han
Differential Privacy Private Federated Learning

November 8, 2022

Privacy-Aware Compression for Federated Learning Through Numerical Mechanism Design
Chuan Guo, Kamalika Chaudhuri, Pierre Stock, Mike Rabbat
Model Compression Mechanism Design Private Federated Learning Privacy Analysis

July 18, 2022

Training Large-Vocabulary Neural Language Models by Private Federated Learning for Resource-Constrained Devices
Mingbin Xu, Congzheng Song, Ye Tian, Neha Agrawal, Filip Granqvist, Rogier van Dalen, Xiao Zhang, Arturo Argueta, Shiyi Han, Yaqiao Deng, Leo Liu, Anmol Walia, Alex Jin
Language Model Differential Privacy Privacy Guarantee Private Federated Learning Resource Constrained Device

June 24, 2022

Using Autoencoders on Differentially Private Federated Learning GANs
Gregor Schram, Rui Wang, Kaitai Liang
Differential Privacy Supervised Autoencoder GAN Model C Gan Private Federated Learning Privacy Sensitive

June 20, 2022

SoteriaFL: A Unified Framework for Private Federated Learning with Communication Compression
Zhize Li, Haoyu Zhao, Boyue Li, Yuejie Chi
Unified Framework Local Differential Privacy Private Federated Learning Communication Compression

May 31, 2022

Private Federated Submodel Learning with Sparsification
Sajani Vithana, Sennur Ulukus
Many Sparse Private Data PRIvacy Leakage Random Sparsification Private Federated Learning

May 6, 2022

Defending against Reconstruction Attacks through Differentially Private Federated Learning for Classification of Heterogeneous Chest X-Ray Data
Joceline Ziegler, Bjarne Pfitzner, Heinrich Schulz, Axel Saalbach, Bert Arnrich
Deep Learning Model Classification Code Differential Privacy Private Federated Learning Reconstruction Attack Chest X Ray Datasets Chest X Ray Classification Non Private

March 31, 2022

Differentially Private Federated Learning via Reconfigurable Intelligent Surface
Yuhan Yang, Yong Zhou, Youlong Wu, Yuanming Shi
Private Federated Learning Air Federated Learning Air Computation Reconfigurable Intelligent Surface Radio Environment

March 15, 2022

Training a Tokenizer for Free with Private Federated Learning
Eugene Bagdasaryan, Congzheng Song, Rogier van Dalen, Matt Seigel, Áine Cahill
Differential Privacy Private Data Private Federated Learning Online Tokenizer

March 13, 2022

Private Non-Convex Federated Learning Without a Trusted Server
Andrew Lowy, Ali Ghafelebashi, Meisam Razaviyayn
Differential Privacy Private Federated Learning Non Convex Loss Function Private Optimization

March 7, 2022

February 18, 2022

November 17, 2021

Differentially Private Federated Learning on Heterogeneous Data
Maxence Noble, Aurélien Bellet, Aymeric Dieuleveut
Differential Privacy Heterogeneous Data Private Federated Learning Privacy Utility Trade

Private Federated Learning

Papers

Adap DP-FL: Differentially Private Federated Learning with Adaptive Noise

Privacy-Aware Compression for Federated Learning Through Numerical Mechanism Design

Training Large-Vocabulary Neural Language Models by Private Federated Learning for Resource-Constrained Devices

Using Autoencoders on Differentially Private Federated Learning GANs

SoteriaFL: A Unified Framework for Private Federated Learning with Communication Compression

Private Federated Submodel Learning with Sparsification

Defending against Reconstruction Attacks through Differentially Private Federated Learning for Classification of Heterogeneous Chest X-Ray Data

Differentially Private Federated Learning via Reconfigurable Intelligent Surface

Training a Tokenizer for Free with Private Federated Learning

Private Non-Convex Federated Learning Without a Trusted Server

The Fundamental Price of Secure Aggregation in Differentially Private Federated Learning

Differentially Private Federated Learning with Local Regularization and Sparsification

FedEmbed: Personalized Private Federated Learning

Differentially Private Federated Learning via Inexact ADMM with Multiple Local Updates

Differentially Private Federated Learning on Heterogeneous Data