Skip to content

Gradient

Jump to bottom
Ilya Gyrdymov edited this page Sep 16, 2017 · 2 revisions

Foundation of gradient optimization methods is a gradient. Gradient is a vector which is consisted of partial derivatives, thus this vector points in the direction of the greatest growth. For example, let us consider the following function:

To the table of contents

Clone this wiki locally