Privacy Preserving Machine Learning

Privacy-preserving machine learning (PPML) aims to enable the development and deployment of machine learning models without compromising the privacy of sensitive training data. Current research heavily focuses on techniques like federated learning (FL), differential privacy (DP), and secure multi-party computation (MPC), often employing model architectures such as neural networks and decision trees, to achieve this goal. These methods address various privacy threats, including membership inference attacks and gradient leakage, with a strong emphasis on balancing privacy guarantees with model accuracy and efficiency. The field's impact spans numerous applications, particularly in healthcare and other sensitive data domains, by enabling collaborative model training and inference while adhering to stringent privacy regulations.

Papers

October 19, 2024

Personalized Federated Learning with Adaptive Feature Aggregation and Knowledge Transfer
Keting Yin, Jiayi Mao
Strong Generalization Federated Learning Knowledge Transfer Personalized Federated Learning Feature Aggregation Privacy Preserving Machine Learning Personalized Model

June 23, 2024

Privacy Preserving Machine Learning for Electronic Health Records using Federated Learning and Differential Privacy
Naif A. Ganadily, Han J. Xia
Differential Privacy Electronic Health Record Clinical Note Patient Data Privacy Preserving Machine Learning

May 29, 2024

Enhancing Security and Privacy in Federated Learning using Update Digests and Voting-Based Defense
Wenjie Li, Kai Fan, Jingyuan Zhang, Hui Li, Wei Yang Bryan Lim, Qiang Yang
Privacy Policy Privacy Preserving Machine Learning Security Measure Incremental Data Party Computation

May 24, 2024

Scaling up the Banded Matrix Factorization Mechanism for Differentially Private ML
Ryan McKenna
Matrix Factorization Privacy Preserving Machine Learning Privacy Amplification Private Machine Learning Noise Optimization

May 23, 2024

A Systematic and Formal Study of the Impact of Local Differential Privacy on Fairness: Preliminary Results
Karima Makhlouf, Tamara Stefanovic, Heber H. Arcolezi, Catuscia Palamidessi
Machine Learning Differential Privacy Procedural Fairness Preliminary Study Local Differential Privacy Local Model Privacy Preserving Machine Learning

April 17, 2024

A Secure and Trustworthy Network Architecture for Federated Learning Healthcare Applications
Antonio Boiano, Marco Di Gennaro, Luca Barbieri, Michele Carminati, Monica Nicoli, Alessandro Redondi, Stefano Savazzi, Albert Sund Aillet, Diogo Reis Santos, Luigi Serio
Federated Learning Privacy Preserving Machine Learning Stroke Prediction

March 26, 2024

Hawk: Accurate and Fast Privacy-Preserving Machine Learning Using Secure Lookup Table Computation
Hamza Saleem, Amir Ziashahabi, Muhammad Naveed, Salman Avestimehr
Machine Learning Differential Privacy Logistic Regression Privacy Preserving Machine Learning Lookup Table

March 17, 2024

Pencil: Private and Extensible Collaborative Learning without the Non-Colluding Assumption
Xuanqi Liu, Zhuotao Liu, Qi Li, Ke Xu, Mingwei Xu
Homomorphic Encryption Collaborative Learning Privacy Preserving Machine Learning Private Model Training Model Privacy Secure Multiparty Computation Tacit Collusion Pen Like Object

February 17, 2024

Neural Networks with (Low-Precision) Polynomial Approximations: New Insights and Techniques for Accuracy Improvement
Chi Zhang, Jingjing Fan, Man Ho Au, Siu Ming Yiu
Neural Network Barzilai Borwein Technique Low Precision Accuracy Improvement Privacy Preserving Machine Learning Polynomial Approximation Privacy Preserving Inference

February 15, 2024

An advanced data fabric architecture leveraging homomorphic encryption and federated learning
Sakib Anwar Rieyan, Md. Raisul Kabir News, A. B. M. Muntasir Rahman, Sadia Afrin Khan, Sultan Tasneem Jawad Zaarif, Md. Golam Rabiul Alam, Mohammad Mehedi Hassan, Michele Ianni, Giancarlo Fortino
LeArning Abstract Medical Image Homomorphic Encryption Privacy Preserving Machine Learning Dataflow Accelerator Federated Learning Architecture

February 14, 2024

I can't see it but I can Fine-tune it: On Encrypted Fine-tuning of Transformers using Fully Homomorphic Encryption
Prajwal Panzade, Daniel Takabi, Zhipeng Cai
Transformer Megatron Decepticons Homomorphic Encryption Privacy Preserving Machine Learning Private Fine Tuning Privacy Preserving Training Encrypted Model

February 5, 2024

On the Impact of Output Perturbation on Fairness in Binary Linear Classification
Vitalii Emelianov, Michaël Perrot
Differential Privacy Global Impact Procedural Fairness Group Fairness Linear Classifier Privacy Preserving Machine Learning Individual Fairness

January 23, 2024

Wasserstein Differential Privacy
Chengyi Yang, Jiayin Qi, Aimin Zhou
Differential Privacy PRIvacy Leakage Privacy Preserving Machine Learning Privacy Budget Privacy Accounting

December 31, 2023

Improving the Privacy and Practicality of Objective Perturbation for Differentially Private Linear Learners
Rachel Redberg, Antti Koskela, Yu-Xiang Wang
Privacy Policy Practical Application Privacy Protection Private Stochastic Gradient Descent Privacy Preserving Machine Learning Privacy Analysis Private Linear

December 24, 2023

An Empirical Study of Efficiency and Privacy of Federated Learning Algorithms
Sofia Zahri, Hajar Bennouri, Ahmed M. Abdelmoniem
High Efficiency Differential Privacy Empirical Study Privacy Policy Secure Aggregation Internet of Thing Network Privacy Preserving Machine Learning Federated Learning Algorithm

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

November 28, 2023

MIA-BAD: An Approach for Enhancing Membership Inference Attack and its Mitigation with Federated Learning
Soumya Banerjee, Sandip Roy, Sayyed Farid Ahamed, Devin Quinn, Marc Vucovich, Dhruv Nandakumar, Kevin Choi, Abdul Rahman, Edward Bowen, Sachin Shetty
Constructive Approach Membership Inference Attack Faulty Negative Mitigation Privacy Preserving Machine Learning Attack Dataset

November 23, 2023

DPSUR: Accelerating Differentially Private Stochastic Gradient Descent Using Selective Update and Release
Jie Fu, Qingqing Ye, Haibo Hu, Zhili Chen, Lulu Wang, Kuncan Wang, Xun Ran
Differential Privacy Stochastic Gradient Descent Private Stochastic Gradient Descent Privacy Preserving Machine Learning Gradient Update Private Model Training Model Update

November 17, 2023

Exploring Machine Learning Models for Federated Learning: A Review of Approaches, Performance, and Limitations
Elaheh Jafarigol, Theodore Trafalis, Talayeh Razzaghi, Mona Zamankhani
Machine Learning Model Constructive Approach System Performance Fundamental Limitation Privacy Preserving Machine Learning Federated Learning Application Learning Based Point

October 23, 2023

B^2SFL: A Bi-level Blockchained Architecture for Secure Federated Learning-based Traffic Prediction
Hao Guo, Collin Meese, Wanxin Li, Chien-Chung Shen, Mark Nejad
Federated Learning Homomorphic Encryption Federated Averaging Privacy Preserving Machine Learning Blockchain Network Blockchain Based Framework

Privacy Preserving Machine Learning

Papers

Personalized Federated Learning with Adaptive Feature Aggregation and Knowledge Transfer

Privacy Preserving Machine Learning for Electronic Health Records using Federated Learning and Differential Privacy

Enhancing Security and Privacy in Federated Learning using Update Digests and Voting-Based Defense

Scaling up the Banded Matrix Factorization Mechanism for Differentially Private ML

A Systematic and Formal Study of the Impact of Local Differential Privacy on Fairness: Preliminary Results

A Secure and Trustworthy Network Architecture for Federated Learning Healthcare Applications

Hawk: Accurate and Fast Privacy-Preserving Machine Learning Using Secure Lookup Table Computation

Pencil: Private and Extensible Collaborative Learning without the Non-Colluding Assumption

Neural Networks with (Low-Precision) Polynomial Approximations: New Insights and Techniques for Accuracy Improvement

An advanced data fabric architecture leveraging homomorphic encryption and federated learning

I can't see it but I can Fine-tune it: On Encrypted Fine-tuning of Transformers using Fully Homomorphic Encryption

On the Impact of Output Perturbation on Fairness in Binary Linear Classification

Wasserstein Differential Privacy

Improving the Privacy and Practicality of Objective Perturbation for Differentially Private Linear Learners

An Empirical Study of Efficiency and Privacy of Federated Learning Algorithms

Hot PATE: Private Aggregation of Distributions for Diverse Task

MIA-BAD: An Approach for Enhancing Membership Inference Attack and its Mitigation with Federated Learning

DPSUR: Accelerating Differentially Private Stochastic Gradient Descent Using Selective Update and Release

Exploring Machine Learning Models for Federated Learning: A Review of Approaches, Performance, and Limitations

B^2SFL: A Bi-level Blockchained Architecture for Secure Federated Learning-based Traffic Prediction