DPDL perturbs cross-gradients with Gaussian noise for DP and applies cosine similarity calibration to support linear speedup in decentralized training on non-IID data.
Differentially Private ADMM for Convex Dis- tributed Learning: Improved Accuracy via Multi-step Approximation,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
DPDL: Towards Differential Privacy Preservation in Decentralized Stochastic Learning on Non-IID Data
DPDL perturbs cross-gradients with Gaussian noise for DP and applies cosine similarity calibration to support linear speedup in decentralized training on non-IID data.