Open access
Sep 2026
CPSGD: differentially private stochastic gradient descent with chaotic post-processing and momentum optimization
Chaos post-processing and momentum-optimized differentially private stochastic gradient descent (CPSGD), which formally proves that each CPSGD step preserves the (ϵ, δ)-differential privacy (DP) guarantee of the underlying Gaussian mechanism.
Zi-Han Zhao, Zhi-Fang Wang
· Engineering Research Express · 0 citations