Prox-ITEM achieves the minimax-optimal distance-to-solution rate among span-based first-order methods for smooth strongly convex composite problems, with Prox-TMM as its stationary limit matching TMM rates.
Taylor and Julien M
4 Pith papers cite this work, alongside 156 external citations. Polarity classification is still indexing.
4
Pith papers citing it
156
external citations · OpenAlex
citation-role summary
other 1
citation-polarity summary
fields
math.OC 4years
2026 4roles
other 1polarities
unclear 1representative citing papers
Establishes matching Ω(T^{-p/(p-1)}) lower bounds for Frank-Wolfe on p-uniformly convex feasible sets for p ≥ 3, plus extension to Hölderian error bounds.
Sparse optimization and SDP-based post-processing of PEP dual certificates recovers compact, interpretable proofs and Lyapunov functions for first-order optimization methods.
Two restart-free accelerated first-order methods for nonconvex functions with Lipschitz gradients and Hessians achieve O(ε^{-7/4}) complexity by discretizing a new ODE model, with adaptive Lipschitz estimation in one variant.
citing papers explorer
-
An optimal first-order method for smooth and strongly convex composite optimization and its stationary limit
Prox-ITEM achieves the minimax-optimal distance-to-solution rate among span-based first-order methods for smooth strongly convex composite problems, with Prox-TMM as its stationary limit matching TMM rates.
-
Curvature-Dependent Lower Bounds for Frank-Wolfe
Establishes matching Ω(T^{-p/(p-1)}) lower bounds for Frank-Wolfe on p-uniformly convex feasible sets for p ≥ 3, plus extension to Hölderian error bounds.
-
Finding Simple Proofs for First-Order Optimization
Sparse optimization and SDP-based post-processing of PEP dual certificates recovers compact, interpretable proofs and Lyapunov functions for first-order optimization methods.
-
A Restart-Free Accelerated Algorithm for Non-Convex Minimization: Continuous and Discrete Analysis
Two restart-free accelerated first-order methods for nonconvex functions with Lipschitz gradients and Hessians achieve O(ε^{-7/4}) complexity by discretizing a new ODE model, with adaptive Lipschitz estimation in one variant.