Gaussian Process Basics

author: David MacKay, University of Cambridge

Description

How on earth can a plain old Gaussian distribution be useful for sophisticated regression and machine learning tasks?

Categories

Top: Computer Science: Machine Learning
Top: Computer Science: Machine Learning: Gaussian Processes

You might be experiencing some problems with Your Video player.

*Slides*
0:02	Gaussian Process Basics
0:55	Nonlinear regression with neural networks
1:24	References
2:00	Motivation: Machine learning
3:00	Can all this be done with a plain old Gaussian distribution?
3:39	Two-dimensional Gaussian
7:03	Inference
7:26	Inference01
8:32	Another representation
11:28	Another representation01
11:45	Another representation02
13:35	Aha!
17:25	Gaussian quiz
24:08	How do we build Gaussian distribution?
25:25	How the matrix was made
27:32	Extend to more points
29:47	A Gaussian process
34:09	Effect of hyperparameters
37:03	Inference of hyperparameters
48:01	Two-dimensional input space
50:09	Efficient computation (... well, modestly efficient)
52:45	Key computational requirements
54:15	Choosing covariance functions
55:51	\"Squared exponential\"
56:10	Emulate infinite neural netwoks
56:29	GPs for classification
58:19	Connection to standard neural networks
60:17	Gaussian Quiz solutions
60:33	Gaussian processes compared with state-of-the-art nonlinear parametric models

Lecture rating

People found this lecture:
Worth seeing
because it is:
Valuable and informative
Well presented
Easily understandable
Acceptably recorded
You need to login to cast your vote.

Report a problem or upload files

If you have found a problem with this lecture or would like to send us extra material, articles, exercises, etc., please use our ticket system to describe your request and upload the data.
Enter your e-mail into the 'Cc' field, and we will keep you updated with your request's status.