Private Aggregation

Private aggregation focuses on enabling collaborative computation and model training on sensitive data while preserving individual privacy. Current research emphasizes techniques like differential privacy, applied to various algorithms including Private Aggregation of Teacher Ensembles (PATE) and its variants, and secure aggregation methods within federated learning frameworks. These advancements are crucial for addressing privacy concerns in diverse applications, ranging from geographical data analysis and medical image segmentation to machine learning model training on sensitive datasets. The field is actively exploring improved privacy-utility trade-offs and mitigating potential biases or vulnerabilities in these aggregation methods.

Papers

October 30, 2024

BUZZ: Beehive-structured Sparse KV Cache with Segmented Heavy Hitters for Efficient LLM Inference
Junqi Zhao, Zhijin Fang, Shu Li, Shaohui Yang, Shichao He
Large Language Model Natural Language Processing LLM Inference Long Document Summarization Key Value Cache Private Aggregation Inference Speedup

October 29, 2024

$\mathsf{OPA}$: One-shot Private Aggregation with Single Client Interaction and its Applications to Federated Learning
Harish Karthikeyan, Antigoni Polychroniadou
Interaction Generation Secure Aggregation Individual Perspective Private Aggregation Secure Computation

June 24, 2024

Learning-Based Heavy Hitters and Flow Frequency Estimation in Streams
Rana Shahout, Michael Mitzenmacher
Link Stream Flow Estimation Private Aggregation Competitive Algorithm Frequency Estimation Partitioned B Bit Hashing

May 6, 2024

Unified Locational Differential Privacy Framework
Aman Priyanshu, Yash Maurya, Suriya Ganesh, Vy Tran
Differential Privacy Private Aggregation Consistent Aggregation

January 15, 2024

Quantum Privacy Aggregation of Teacher Ensembles (QPATE) for Privacy-preserving Quantum Machine Learning
William Watkins, Heehwan Wang, Sangyoon Bae, Huan-Hsin Tseng, Jiook Cha, Samuel Yen-Chi Chen, Shinjae Yoo
Machine Learning Quantum Machine Learning Quantum Physic Quantum Neural Network Quantum Machine Private Aggregation Ensemble Teacher

December 4, 2023

Hot PATE: Private Aggregation of Distributions for Diverse Task
Edith Cohen, Benjamin Cohen-Wang, Xin Lyu, Jelani Nelson, Tamas Sarlos, Uri Stemmer
Generative Model Sensitive Data Privacy Preserving Machine Learning Autoregressive Generative Model Diverse Task Private Aggregation

August 1, 2023

Differential Privacy for Adaptive Weight Aggregation in Federated Tumor Segmentation
Muhammad Irfan Khan, Esa Alhoniemi, Elina Kontio, Suleiman A. Khan, Mojtaba Jafaritadi
Differential Privacy Weight Aggregation Private Aggregation

July 21, 2023

Differentially Private Heavy Hitter Detection using Federated Analytics
Karan Chadha, Junye Chen, John Duchi, Vitaly Feldman, Hanieh Hashemi, Omid Javidbakht, Audra McMillan, Kunal Talwar
Local Differential Privacy Privacy Constraint Federated Data Field Data Private Aggregation

June 25, 2023

Private Aggregation in Wireless Federated Learning with Heterogeneous Clusters
Maximilian Egger, Christoph Hofmeister, Antonia Wachter-Zeh, Rawad Bitar
Federated Learning Gradient Descent Privacy Issue Private Aggregation Heterogeneous Cluster Information Theoretic Privacy

May 24, 2023

Differentially-Private Decision Trees and Provable Robustness to Data Poisoning
Daniël Vos, Jelle Vos, Tianyu Li, Zekeriya Erkin, Sicco Verwer
Differential Privacy Decision Tree Data Poisoning Private Aggregation

May 19, 2023

On the Fairness Impacts of Private Ensembles Models
Cuong Tran, Ferdinando Fioretto
Differential Privacy Semi Supervised Private Model Private Aggregation Fairness Implication Ensemble Teacher Private Ensemble

May 9, 2023

FedGT: Identification of Malicious Clients in Federated Learning with Secure Aggregation
Marvin Xhemrishi, Johan Östman, Antonia Wachter-Zeh, Alexandre Graell i Amat
Person Identification Secure Aggregation Malicious Client Private Aggregation

April 20, 2023

DPAF: Image Synthesis via Differentially Private Aggregation in Forward Phase
Chih-Hsun Lin, Chia-Yi Hsu, Chia-Mu Yu, Yang Cao, Chun-Ying Huang
Synthetic Data Image Synthesis Private Stochastic Gradient Descent High Resolution Image Synthesis Civil Engineering Phase Private Generative Model Private Aggregation

December 31, 2022

Generalized PTR: User-Friendly Recipes for Data-Adaptive Algorithms with Differential Privacy
Rachel Redberg, Yuqing Zhu, Yu-Xiang Wang
Differential Privacy Complete Recipe Dependent Data Private Aggregation Private Linear Regression

November 3, 2022

Private Semi-supervised Knowledge Transfer for Deep Learning from Noisy Labels
Qiuchen Zhang, Jing Ma, Jian Lou, Li Xiong, Xiaoqian Jiang
Deep Learning Generative Adversarial Network Noisy Label Private Aggregation Noisy Label Detection

October 11, 2022

An Experimental Study on Private Aggregation of Teacher Ensemble Learning for End-to-End Speech Recognition
Chao-Han Huck Yang, I-Fan Chen, Andreas Stolcke, Sabato Marco Siniscalchi, Chin-Hui Lee
Differential Privacy End to End Experimental Study Private Aggregation Acoustic Side Channel Ensemble Teacher

September 22, 2022

In Differential Privacy, There is Truth: On Vote Leakage in Ensemble Private Learning
Jiaqi Wang, Roei Schuster, Ilia Shumailov, David Lie, Nicolas Papernot
Differential Privacy Differential Privacy Guarantee Great Truth Private Aggregation Privacy Preserving Training Election Manipulation Private Ensemble

September 12, 2022

Cocktail Party Attack: Breaking Aggregation-Based Privacy in Federated Learning using Independent Component Analysis
Sanjay Kariyappa, Chuan Guo, Kiwan Maeng, Wenjie Xiong, G. Edward Suh, Moinuddin K Qureshi, Hsien-Hsin S. Lee
Independent Component Analysis Gradient Inversion Attack Blind Source Separation Gradient Inversion Private Aggregation Cocktail Party

July 3, 2022

Tricking the Hashing Trick: A Tight Lower Bound on the Robustness of CountSketch to Adaptive Inputs
Edith Cohen, Jelani Nelson, Tamás Sarlós, Uri Stemmer
Native Robustness Lower Bound Adversarial Input Private Aggregation Input Adaptation Partitioned B Bit Hashing

April 11, 2022

SF-PATE: Scalable, Fair, and Private Aggregation of Teacher Ensembles
Cuong Tran, Keyu Zhu, Ferdinando Fioretto, Pascal Van Hentenryck
Neural Network Potential Scalability Fair Model Private Aggregation Ensemble Teacher

Private Aggregation

Papers

BUZZ: Beehive-structured Sparse KV Cache with Segmented Heavy Hitters for Efficient LLM Inference

$\mathsf{OPA}$: One-shot Private Aggregation with Single Client Interaction and its Applications to Federated Learning

Learning-Based Heavy Hitters and Flow Frequency Estimation in Streams

Unified Locational Differential Privacy Framework

Quantum Privacy Aggregation of Teacher Ensembles (QPATE) for Privacy-preserving Quantum Machine Learning

Hot PATE: Private Aggregation of Distributions for Diverse Task

Differential Privacy for Adaptive Weight Aggregation in Federated Tumor Segmentation

Differentially Private Heavy Hitter Detection using Federated Analytics

Private Aggregation in Wireless Federated Learning with Heterogeneous Clusters

Differentially-Private Decision Trees and Provable Robustness to Data Poisoning

On the Fairness Impacts of Private Ensembles Models

FedGT: Identification of Malicious Clients in Federated Learning with Secure Aggregation

DPAF: Image Synthesis via Differentially Private Aggregation in Forward Phase

Generalized PTR: User-Friendly Recipes for Data-Adaptive Algorithms with Differential Privacy

Private Semi-supervised Knowledge Transfer for Deep Learning from Noisy Labels

An Experimental Study on Private Aggregation of Teacher Ensemble Learning for End-to-End Speech Recognition

In Differential Privacy, There is Truth: On Vote Leakage in Ensemble Private Learning

Cocktail Party Attack: Breaking Aggregation-Based Privacy in Federated Learning using Independent Component Analysis

Tricking the Hashing Trick: A Tight Lower Bound on the Robustness of CountSketch to Adaptive Inputs

SF-PATE: Scalable, Fair, and Private Aggregation of Teacher Ensembles