Examples of Pathological Dynamics of the Subgradient Method for Lipschitz Path-Differentiable Functions

DOI10.1287/moor.2021.1241OpenAlexW4229027915MaRDI QIDQ5870372

Rodolfo Ríos-Zertuche

Publication date: 9 January 2023

Published in: Mathematics of Operations Research (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2007.11699

zbMATH Keywords

large-scale optimizationgradient descent

Mathematics Subject Classification ID

Numerical optimization and variational techniques (65K10) Dynamical systems and their relations with probability theory and stochastic processes (37A50) Gradient-like behavior; isolated (locally maximal) invariant sets; attractors, repellers for topological dynamical systems (37B35)

Related Items

Conservative parametric optimality and the ridge method for tame min-max problems

Cites Work

• Unnamed Item
• Action minimizing invariant measures for positive definite Lagrangian systems
• Convergence of descent methods for semi-algebraic and tame problems: proximal algorithms, forward-backward splitting, and regularized Gauss-Seidel methods
• Conservative set valued fields, automatic differentiation, stochastic gradient methods and deep learning
• Stochastic subgradient method converges on tame functions
• A function not constant on a connected set of critical points
• Generalized subdifferentials: a Baire categorical approach
• Introduction to Piecewise Differentiable Equations
• Characterizations of Łojasiewicz inequalities: Subgradient flows, talweg, convexity
• Differentiable functions
• A Chain Rule for Essentially Smooth Lipschitz Functions
• Pathological Subgradient Dynamics