Characterization of optimization problems that are solvable iteratively with linear convergence

2024/02/19 by Alimisis, Foivos · 1 citation
#Differential Geometry (math.DG) #FOS: Mathematics #Optimization and Control (math.OC)

paper · doi:10.48550/arxiv.2402.12090

Abstract

In this work, we state a general conjecture on the solvability of optimization problems via algorithms with linear convergence guarantees. We make a first step towards examining its correctness by fully characterizing the problems that are solvable via Riemannian gradient descent with linear convergence.

Cited by

• Continuized Nesterov Acceleration for Non-Convex Optimization

Related

• Asymptotic convergence of iterative optimization algorithms
• Learning to optimize with guarantees: a complete characterization of linearly convergent algorithms
• Iterated Linear Optimization
• Convergence of Successive Linear Programming Algorithms for Noisy Functions
• Convergence and Complexity Guarantee for Inexact First-order Riemannian Optimization Algorithms