Matching Full-Hessian Oracle Complexity for Nonconvex--Strongly-Concave Minimax Optimization
Year: 2026
Type: preprint
Abstract: We characterize the oracle complexity of finding first-order stationary points in smooth nonconvex--strongly-concave (NC--SC) minimax optimization when each query provides the full Hessian in addition to the function value and gradient. For $Φ(x)=\max_y f(x,y)$, assume that $f$ has an $L$-Lipschitz gradient, a $ρ$-Lipschitz Hessian, and is $μ$-strongly concave in $y$. Let $κ=L/μ$, let $Δ$ denote the initial value gap, and define $r=ρκ^{3/2}\varepsilon/L^2$. We establish matching deterministic upper bounds and randomized lower bounds, up to logarithmic factors and additive initialization costs. When the leading term dominates initialization, the resulting minimax complexity is \[ \widetildeΘ\!\left( \frac{ΔL}{\varepsilon^2}Ψ_κ(r) \right), \qquad Ψ_κ(r)= \begin{cases} r^{1/2}, & 0
Source: arXiv (Cornell University)
Institution: Peking University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7221803062