12 lines (9 loc) · 1.01 KB

Gradient Methods

Demonstration of gradient descent methods

Description of files

The script Gradient_Descent.m solves the linear system $y \= Hx$ using the steepest gradient descent method, where:
$H \in \mathbb{R}^{m \times n}$ , $x \in \mathbb{R}^{n \times 1}$ and $y \in \mathbb{R}^{m \times 1}$

Visualization

The script Gradient_Visualize.m solves the same system in 2 dimensions and plots the trajectory of the solution vector ( in red ) in the solution search space at each iteration.
Eventually, the solution vector converges to the expected analytical solution ( in green )