DP- lambda CGD: Efficient Noise Correlation for Differentially Private Model Training
cs.LG
Submitted: 2026-01-29
Updated: 2026-09-26
Code: https://github.com/google-research/federated
Terminology
Sources
- Efficient and Scalable Implementation of Differentially Private Deep Learning without Shortcuts
- Privacy Amplification for BandMF via $b$-Min-Sep Subsampling
- Privacy amplification by random allocation
- Efficient privacy loss accounting for subsampling and random allocation
- Correlating Cross-Iteration Noise for DP-SGD using Model Curvature
- Normalized Square Root: Sharper Matrix Factorization Bounds for Differentially Private Continual Counting
- Learning Rate Scheduling with Matrix Factorization for Private Training
- Back to Square Roots: An Optimal Bound on the Matrix Factorization Error for Multi-Epoch Differentially Private SGD
- Cocoon: A System Architecture for Differentially Private Training with Correlated Noises
- Correlated Noise Mechanisms for Differentially Private Learning
- Sampling-Free Privacy Accounting for Matrix Mechanisms under Random Allocation
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks