Can I read Global Convergence of Gradient Descent for Phase Retrieval on EtoBox?

Global Convergence of Gradient Descent for Phase Retrieval by Fougereux, Théodore; Josz, Cédric; Li, Xiaopeng is a scholarly article available to read on EtoBox.

What is Global Convergence of Gradient Descent for Phase Retrieval about?

We propose a tensor-based criterion for benign landscape in phase retrieval and establish boundedness of gradient trajectories. This implies that gradient descent will converge to a global minimum for almost every initial point.

Author

Fougereux, Théodore; Josz, Cédric; Li, Xiaopeng

Published

2024

Language

EN