Feb 2024· International Conference on Machine Learning· pp. 840-878· 12 citations· ⚡ 2 influential· 69 references
MathematicsComputer Science
TL;DR
This work provides a refined analysis for inexact Halpern iteration that relaxes the required inexactness level to improve some state-of-the-art complexity results even for constrained stochastic convex-concave min-max problems.
Abstract
We focus on constrained, $L$-smooth, potentially stochastic and nonconvex-nonconcave min-max problems either satisfying $\rho$-cohypomonotonicity or admitting a solution to the $\rho$-weakly Minty Variational Inequality (MVI), where larger values of the parameter $\rho>0$ correspond to a greater degree of nonconvexity. These problem classes include examples in two player reinforcement learning, interaction dominant min-max problems, and certain synthetic test problems on which classical min-max algorithms fail. It has been conjectured that first-order methods can tolerate a value of $\rho$ no larger than $\frac{1}{L}$, but existing results in the literature have stagnated at the tighter requirement $\rho<\frac{1}{2L}$. With a simple argument, we obtain optimal or best-known complexity guarantees with cohypomonotonicity or weak MVI conditions for $\rho<\frac{1}{L}$. First main insight for the improvements in the convergence analyses is to harness the recently proposed $\textit{conic nonexpansiveness}$ property of operators. Second, we provide a refined analysis for inexact Halpern iteration that relaxes the required inexactness level to improve some state-of-the-art complexity results even for constrained stochastic convex-concave min-max problems. Third, we analyze a stochastic inexact Krasnosel'ski\u{\i}-Mann iteration with a multilevel Monte Carlo estimator when the assumptions only hold with respect to a solution.
We study whether the linear condition-number dependence in the stochastic complexity of SAPD+ is necessary for nonconvex-strongly-concave minimax optimization. For jointly $L$-smooth objectives with dual strong-concavity parameter $\mu$, we prove a lower bound that matches the SAPD+ upper bound under the same Moreau-en...
How far can stochastic gradient descent ascent (SGDA) go by tuning its timescale ratio and step sizes in nonconvex min-max games? We answer this question for nonconvex-P{\L} (NC-P{\L}) games by establishing the first tight complexity of two-timescale SGDA with a fixed timescale ratio and non-increasing step sizes. For...
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 $\Phi(x)=\max_y f(x,y)$, assume that $f$ has an $L$-Lipschitz gradient, a $\rho...
Fang-Qing Zhang, Hai-Han Zhang, Wen-Dao Wu et al.· 0 citations
We study the sample complexity of finding a point with expected hypergradient norm at most $\epsilon>0$ in nonconvex--strongly-convex stochastic bilevel optimization using fresh, globally unbiased first-order samples with bounded variance. For lower-variable smoothness order $p\ge1$, F$^2$SA-$p$ achieves $\widetilde O(...
Wen-Dao Wu, Hai-Han Zhang, Chen-Heng Zhang et al.· 0 citations
We study lower bounds on the first-order oracle complexity of smooth nonconvex-concave minimax optimization. We consider objectives $f$ that are jointly $L$-smooth in the primal and dual variables $(x,y)$, concave in $y$, and whose primal value function $\Phi(x) := \max_{y\in\mathcal Y} f(x,y)$ satisfies the initial-ga...
A tight lower bound is proved of $\Omega(\lambda^{1/2}\epsilon^{-7/2})$ for finding points satisfying the recently proposed relaxed notion of $(\lambda,\epsilon)$-stationarity, which allows combining further-away gradients.
Guy Kornowski· 3 citations
Related blog posts
MIT News · Artificial Intelligence· news.mit.eduNov 6, 2026
A machine-learning approach called CheMeleon provides a practical way to develop new drugs or other molecules with much less detailed, expensive experimentation.
Learn how Google DeepMind and GovTech Singapore uses lightweight on-device Gemma models to intercept conversational scams in private chats with zero cloud data egress.