Location: EU Supported » PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning » Machine Learning Summer School (MLSS), Cambridge 2009

Learning Theory

author: John Shawe-Taylor, Centre for Computational Statistics and Machine Learning, University College London
published: Nov. 2, 2009,   recorded: September 2009,   views: 1442
Categories
You might be experiencing some problems with Your Video player.

Slides

Slides
0:00 Learning Theory
5:38 Theories of learning
10:18 Theories of learning cont.
13:26 General statistical considerations
13:46 General statistical considerations cont.
15:58 Generalisation of a learner (1)
16:26 Generalisation of a learner (2)
17:52 Example of Generalisation I
19:00 Example of Generalisation II
20:28 Example of Generalisation III
21:52 Error distribution: full dataset
22:04 Error distribution: dataset size: 342
22:48 Error distribution: dataset size: 273
22:52 Error distribution: dataset size: 205
23:00 Error distribution: dataset size: 137
23:16 Error distribution: dataset size: 68
23:28 Error distribution: dataset size: 34
23:40 Error distribution: dataset size: 27
23:48 Error distribution: dataset size: 20
23:56 Error distribution: dataset size: 14
24:00 Error distribution: dataset size: 7
24:04 Bayes risk and consistency
26:28 Error distribution: dataset size: 68
26:40 Expected versus confident bounds cont.
29:16 Expected versus confident bounds cont.
30:32 Probability of being misled in classification
32:56 Finite or Countable function classes
36:14 Finite or Countable function classes result
38:38 Some comments on the result
40:46 What if uncountably many functions?
48:34 Finite or Countable function classes result
55:52 What if uncountably many functions?
57:03 Double sample trick
59:26 Double sample trick II (1)
60:14 Double sample trick II (2)
60:58 How many functions on a finite sample?
61:42 Examining the growth function
62:42 Vapnik Chervonenkis dimension
65:16 Sauer’s Lemma
67:00 Basic Theorem of SLT
68:04 Symmetrisation
69:04 Symmetrisation cont.
70:00 Completion of the proof
70:44 Final result
72:28 Lower bounds
73:28 Non-zero training error
73:48 Structural Risk Minimisation
76:48 Criticisms of PAC Theory
78:20 Support Vector Machines cont.
78:35 Margin in SVMs
80:07 Margin in SVMs cont.
80:37 Learning Theory

Related content

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.
Lecture popularity: You need to login to cast your vote.
 
    Delicious Bibliography

 Watch videos:   (click on thumbnail to launch)

Watch Part 1
Part 1 1:21:00
!NOW PLAYING
Watch Part 2
Part 2 1:23:42

See Also:

Launch Windows Media PlayerLaunch in a standalone WM Player

WMedia PlayerSwitch to Windows Media Player

Download slides icon Download slides: mlss09uk_shawe_taylor_lt.pdf (299.2 KB)

Help icon Streaming Video Help

WebLink icon Windows Media Player Firefox Plugin - Download

Link this page

Would you like to put a link to this lecture on your homepage?
Go ahead! Copy the HTML snippet !

Reviews and comments:

Comment1 Ed Warner, January 6, 2011 at 12:43 p.m.:

I enjoyed this very competent talk by this very likable speaker. Dr. Shawe-Taylor says that he views Learning Theory as much broader than just Statistical Learning Theory, which is his focus. It would have been nice, however, to have another talk on Algorithmic Learning Theory at this summer school, as one could argue that ALT is the most general type of learning theory.

Write your own review or comment:

make sure you have javascript enabled or clear this field: