Exponentially-shifted Gaussian smoothing yields zeroth-order gradient estimators with linear dimension dependence, enabling improved complexity bounds for stochastic optimization including decision-dependent regimes.
Stochastic optimization schemes for performative prediction with nonconvex loss.Advances in Neural Information Processing Systems, 37:8673–8697, 2024
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
math.OC 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Complexity Guarantees for Zeroth-order Methods via Exponentially-shifted Gaussian Smoothing: Mitigating Dimension-dependence and Incorporating Decision-dependence
Exponentially-shifted Gaussian smoothing yields zeroth-order gradient estimators with linear dimension dependence, enabling improved complexity bounds for stochastic optimization including decision-dependent regimes.