3 papers
stat.ME2026
Near-Oracle Robustification of Finite-Difference Stochastic Gradient Estimators via Cheap Pilot Calibration
Haidong Li, Henry Lam, Yijie Peng
We study stochastic gradient estimation in black-box environments where only noisy simulation observations of function values are available. Finite-difference (FD) methods are amon…
math.NT2026
Large values of logarithmic derivatives of quadratic Dirichlet -functions
Zikang Dong, Haidong Li
In this article, we apply the resonance method to derive conditional Omega results for logarithmic derivatives of quadratic Dirichlet -functions. We improve a previous result of…
math.ST2020
Minimax Efficient Finite-Difference Stochastic Gradient Estimators Using Black-Box Function Evaluations
Henry Lam, Haidong Li, Xuhui Zhang
Standard approaches to stochastic gradient estimation, with only noisy black-box function evaluations, use the finite-difference method or its variants. While natural, it is open t…