19. Saddle Points Continued, Maxmin Principle
2.71M subscribers
SUBSCRIBED
MIT 18.065 Matrix Methods in Data Analysis, Signal Processing, and Machine Learning, Spring 2018 Instructor: Gilbert Strang View the complete course: https://ocw.mit.edu/18-065S18YouTube Playlist: https://www.youtube.com/playlist?list...Professor Strang continues his discussion of saddle points, which are critical for deep learning applications. Later in the lecture, he reviews the Maxmin Principle, a decision rule used in probability and statistics to optimize outcomes. License: Creative Commons BY-NC-SA More information at https://ocw.mit.edu/termsMore courses at https://ocw.mit.edu
00:00
00:01
The following content is provided under a Creative
00:03
Commons license.
00:05
Your support will help MIT OpenCourseWare
00:07
continue to offer high quality educational resources for free.
00:11
To make a donation or to view additional materials
00:14
from hundreds of MIT courses, visit MIT OpenCourseWare
00:18
at ocw.mit.edu.
00:19
00:23
GILBERT STRANG: So I've got a list of things
00:26
I'm hoping to do today.
00:28
I'll begin with a few final words about saddle points.
00:32
The reason I'm interested in saddle points
00:35
is when we get to this deep learning direction,
00:42
you know that the big step there is
00:45
finding a minimum of the total cost function and gradient
00:53
descent, which we'll certainly discuss
00:55
as the usual method or stochastic gradient descent.
01:00
And all kinds of issues arise, what
01:04
happens if you have a saddle point or a degenerate minimum?
01:11
All these possibilities and the understanding of deep learning
01:17
is focusing more and more on what