video thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Gradient Methods for Machine Learning

Published on 2007-02-259994 Views

Gradient methods locally optimize an unknown differentiable function, and thus provide the engines that drive much machine learning. Here we'll take a look under the hood, beginning with brief overvie

Related categories

Presentation

Course Overview00:01
Classical Gradient Methods01:34
Function Optimization04:01
Methods by Gradient Order06:46
Direct (Gradient-Free) Methods08:22
Prototypical Direct Method09:25
Direct Methods: Advantages13:23
Direct Methods: Disadvantages18:33
Gradient Descent21:42
Gradient Descent: Disadvantages25:08
Newton’s Method31:52
Gauss-Newton Approximation38:32
Levenberg-Marquardt48:09
Quasi-Newton: BFGS50:42
Conjugate Gradient53:15
Conjugate Gradient: Properties57:36