DP-GD achieves minimax optimal non-asymptotic risk O(γ + γ²/ρ²) for well-conditioned high-dimensional data and power-law scaling for ill-conditioned power-law spectra, with the exponent depending on the privacy parameter ρ.
Hitting the high- dimensional notes: an ode for sgd learning dynamics on glms and multi-index models.Information and Inference: A Journal of the IMA, 13(4):iaae028, 2024a
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
representative citing papers
In the high-dimensional regime, SGD on diagonal linear networks is approximated by an SDE and a deterministic PDE that together give an explicit non-asymptotic description of convergence to zero risk.
citing papers explorer
-
High-Dimensional Private Linear Regression with Optimal Rates
DP-GD achieves minimax optimal non-asymptotic risk O(γ + γ²/ρ²) for well-conditioned high-dimensional data and power-law scaling for ill-conditioned power-law spectra, with the exponent depending on the privacy parameter ρ.
-
High-dimensional Limit of SGD for Diagonal Linear Networks
In the high-dimensional regime, SGD on diagonal linear networks is approximated by an SDE and a deterministic PDE that together give an explicit non-asymptotic description of convergence to zero risk.