Robustly stable accelerated momentum methods with a near-optimal L₂ gain and H∞ performance

Mathematics of Operations Research, 2025. Dedicated to Michael L. Overton.

Publisher (DOI)

Treats first-order algorithms as feedback dynamical systems subject to worst-case, finite-energy gradient errors and computes their H∞ norm in closed form on quadratics — together with the explicit worst-case noise achieving it (a decaying cosine). Introduces robustly stable variants (RS-GD, RS-HB) with near-optimal L₂ gain, quantifies real stability radii of momentum methods, and extends the guarantees beyond quadratics through matrix-inequality certificates.

The playground The speed–robustness trade-off lets you try the ideas of this paper in the browser.

The research note How fragile is acceleration? explains the paper informally.

Momentum & accelerationRisk & robustnessSpectral & nonsmooth optimization

• Accelerated gradient methods for nonconvex optimization: escape trajectories from strict saddle points and convergence to local minima (Found. Comput. Math., 2026)
• Accelerated gradient methods with biased gradient estimates: risk sensitivity, high-probability guarantees, and large deviation bounds (J. Nonlinear Var. Anal., 2026)
• Algorithmic stability of stochastic gradient descent with momentum under heavy-tailed noise (arXiv, 2025)
• Entropic risk-averse generalized momentum methods (Optim. Methods Softw., 2025)
• Robust accelerated primal-dual methods for computing saddle points (SIAM J. Optim., 2024)