November 18, 2025Numerical AlgorithmsOpen Access

The essential best and average rate of convergence of the exact line search gradient descent method

View Full Paper

Ask AI

Bookmark

Authors

Discussion

Analysis reveals the average rate of convergence in the phase retrieval problem, suggesting faster convergence in ill-conditioned scenarios.

Average rate of convergence improves in ill-conditioned settings, highlighting intriguing dynamics.
Key metric: average ROC approaches worst-case ROC in convex quadratic objectives as condition increases.
Utilizing a dynamical system approach reveals the essential best case rate of convergence.
Analysis may enable optimized algorithms in polynomial optimization problems like phase retrieval.

Thomas Yu (2025) studied this question.

View Full Paper

Ask AI

Bookmark

Analysis reveals the average rate of convergence in the phase retrieval problem, suggesting faster convergence in ill-conditioned scenarios.