Differential Privacy

Differential privacy (DP) is a rigorous framework for ensuring data privacy in machine learning by adding carefully calibrated noise to model training processes. Current research focuses on improving the accuracy of DP models, particularly for large-scale training, through techniques like adaptive noise allocation, Kalman filtering for noise reduction, and novel gradient processing methods. This active area of research is crucial for enabling the responsible use of sensitive data in various applications, ranging from healthcare and finance to natural language processing and smart grids, while maintaining strong privacy guarantees.

Papers

June 6, 2022

Towards Practical Differential Privacy in Data Analysis: Understanding the Effect of Epsilon on Utility in Private ERM
Yuzhe Li, Yong Liu, Bo Li, Weiping Wang, Nan Liu
Differential Privacy Task Utility Privacy Guarantee Estimation Performance High Probability Guarantee Private Empirical Risk Minimization

June 3, 2022

Differentially Private Model Compression
Fatemehsadat Mireshghallah, Arturs Backurs, Huseyin A Inan, Lukas Wutschitz, Janardhan Kulkarni
Differential Privacy BERT Model Model Compression Large Pre Trained Language Model Non Private

June 2, 2022

June 1, 2022

May 31, 2022

FedWalk: Communication Efficient Federated Unsupervised Node Embedding with Differential Privacy
Qiying Pan, Yifei Zhu
Differential Privacy Graph Representation Unsupervised Federated Learning Visibility Graph

May 28, 2022

May 27, 2022

May 26, 2022

Privacy-Preserving Wavelet Neural Network with Fully Homomorphic Encryption
Syed Imtiaz Ahamed, Vadlamani Ravi
Machine Learning Differential Privacy Homomorphic Encryption Wavelet Neural Network Encrypted Model

May 24, 2022

Differentially Private AUC Computation in Vertical Federated Learning
Jiankai Sun, Xin Yang, Yuanshun Yao, Junyuan Xie, Di Wu, Chong Wang
Differential Privacy Vertical Federated Learning Privacy Preservation Privacy Preserving Computation Privacy Label

May 23, 2022

PrivFairFL: Privacy-Preserving Group Fairness in Federated Learning
Sikha Pentyala, Nicola Neophytou, Anderson Nascimento, Martine De Cock, Golnoosh Farnadi
Differential Privacy Group Fairness Fair Machine Learning Secure Multiparty Computation

May 21, 2022

Scalable and Efficient Training of Large Convolutional Neural Networks with Differential Privacy
Zhiqi Bu, Jialin Mao, Shiyun Xu
Convolutional Neural Network Differential Privacy Potential Scalability Convolutional Layer Efficient Training Differentially Private Clipped Stochastic Gradient Descent High Dimensional Clipping

May 19, 2022

Differential Privacy: What is all the noise about?
Roxana Danger
Machine Learning Differential Privacy Industrial Disturbing Noise Privacy Breach

May 18, 2022

Federated learning: Applications, challenges and future directions
Subrato Bharati, M. Rubaiyat Hossain Mondal, Prajoy Podder, V. B. Surya Prasath
Technical Challenge Financial Application Differential Privacy Stochastic Gradient Descent Future Direction Homomorphic Encryption Global Deep

May 13, 2022

On the Importance of Architecture and Feature Selection in Differentially Private Machine Learning
Wenxuan Bao, Luke A. Bauer, Vincent Bindschaedler
Differential Privacy Feature Selection Privacy Preserving Importance Aware Architecture Design Private Machine Learning Private Learning Algorithm

May 12, 2022

Fair NLP Models with Differentially Private Text Encoders
Gaurav Maheshwari, Pascal Denis, Mikaela Keller, Aurélien Bellet
Differential Privacy Adversarial Training Text Representation Sensitive Attribute Private Training NLP Fairness

May 10, 2022

Sentence-level Privacy for Document Embeddings
Casey Meehan, Khalil Mrini, Kamalika Chaudhuri
Differential Privacy Privacy Guarantee Local Differential Privacy Document Embeddings