AI RESEARCH

Revisiting Zeroth-Order Hessian Approximation: A Single-Step Policy Optimization Lens

arXiv CS.LG

ArXi:2605.30960v1 Announce Type: new Accurate Zeroth-Order (ZO) Hessian estimation is a cornerstone of derivative-free methods, essential for tasks such as bilevel optimization, Bayesian inference, and uncertainty quantification. However, obtaining a complete suite of low-variance estimators for the Hessian and its inverse in high-dimensional settings remains a significant challenge. To address this, we propose a unified framework that reinterprets ZO Hessian approximation through the lens of single-step Policy Optimization (