3 papers
cs.LG2026
Less Random, More Private: What is the Optimal Subsampling Scheme for DP-SGD?
Andy Dong, Ayfer Ãzgür
Poisson subsampling is the default sampling scheme in differentially private machine learning, largely because its unstructured randomness yields tractable privacy amplification an…
cs.LG2025
Leveraging Randomness in Model and Data Partitioning for Privacy Amplification
Andy Dong, Wei-Ning Chen, Ayfer Ozgur
We study how inherent randomness in the training process -- where each sample (or client in federated learning) contributes only to a randomly selected portion of training -- can b…
cs.CR2024
Universal Exact Compression of Differentially Private Mechanisms
Yanxiao Liu, Wei-Ning Chen, Ayfer Ãzgür +1
To reduce the communication cost of differential privacy mechanisms, we introduce a novel construction, called Poisson private representation (PPR), designed to compress and simula…