Year: 2025
Type: preprint
Abstract: We analyze the behavior of randomized coordinate gradient descent for nonconvex optimization, proving that under standard assumptions, the iterates almost surely escape strict saddle points. By formulating the method as a nonlinear random dynamical system and characterizing neighborhoods of critical points, we establish this result through the center-stable manifold theorem.
Source: arXiv (Cornell University)
Authors: Chen, Ziang, Li, Yingzhou, Li, Zihao
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4416242551
Cites:
Cited by:
Related to: