Machine Learning Summer School on Theory and Practice of Computational Learning

Kernel Methods and Support Vector Machines

author: John Shawe-Taylor, Centre for Computational Statistics and Machine Learning, University College London

Description

Kernel methods have become a standard tool for pattern analysis during the last fifteen years since the introduction of support vector machines. We will introduce the key ideas and indicate how this approach to pattern analysis enables a relatively easy plug and play application of different tools. The problem of choosing and designing a kernel for specific types of data will also be considered and an overview of different kernels will be given.

You might be experiencing some problems with Your Video player.

*Slides*
0:00	- Introduction
0:18	Kernel Methods and Support Vector Machines
2:03	Aim:
2:31	What won’t be included:
3:09	OVERALL STRUCTURE
4:24	PART 1 STRUCTURE
4:58	Pattern Analysis
6:45	Defining patterns
7:15	Pattern analysis algorithms
9:01	Brief Historical Perspective
11:01	Kernel methods
13:45	Kernel methods approach
14:46	Kernel methods embedding
15:49	Worked example: Ridge Regression
17:24	Possible pattern function
18:19	Worked example: Ridge Regression
18:21	Possible pattern function
19:15	Optimising the choice of g
19:57	Possible pattern function
20:04	Optimising the choice of g
21:18	Primal solution
21:46	Optimising the choice of g
21:55	Primal solution
22:50	Dual solution (1)
22:56	Primal solution
22:59	Dual solution (1)
24:58	Optimising the choice of g
25:35	Dual solution (1)
27:18	Dual solution (2)
27:26	Dual solution (1)
27:33	Dual solution (2)
27:47	Primal solution
27:49	Optimising the choice of g
28:07	Dual solution (2)
30:31	Key ingredients of dual solution
32:16	Applying the ‘kernel trick’
32:37	Key ingredients of dual solution
33:02	Applying the ‘kernel trick’
33:32	A simple kernel example
36:33	Implications of the kernel trick
38:03	Dual solution (2)
38:39	Implications of the kernel trick
38:53	Implications of kernel algorithms
41:52	Defining kernels
44:44	Means and distances (1)
46:51	Means and distances (2)
48:08	Means and distances (3)
48:38	Means and distances (2)
48:40	Means and distances (1)
48:51	Means and distances (3)
48:56	Means and distances (4)
51:21	Means and distances (5)
51:50	Means and distances (4)
52:10	Means and distances (5)
52:25	Means and distances (6)
53:40	Simple novelty detection

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.