Location: EU Supported » PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning » Machine Learning Summer School (MLSS), Tübingen 2003

Universal Modeling: Introduction to modern MDL

author: Peter Grünwald, Center for Mathematics and Computer Science - CWI
published: Feb. 25, 2007,   recorded: August 2003,   views: 913
Categories
You might be experiencing some problems with Your Video player.

Slides

Slides
0:00 Universal Modeling: Introduction to ‘Modern’ MDL
1:28 Minimum Description Length Principle
2:04 Minimum Description Length Principle
5:51 Minimum Description Length Principle
6:19 Model Selection
7:35 Example: Regression
7:54 Example: Regression
8:23 Example: Regression
8:35 Example: Regression
8:44 Example: Regression
8:46 Modern MDL!
9:40 Five MDL Lectures
10:23 Part I: Overview
10:56 Codes
12:13 Example 1: uniform code
12:54 Code Length & Probability
14:07 Code Lengths ‘are’ probabilities…
15:04 …and probabilities ‘are’ code lengths!
15:40 The Most Important Slide!
16:14 The Most Important Slide!
17:09 Example 1: uniform code/distr.
18:05 Prefix codes distributions
19:41 Prefix codes distributions
20:59 General Recipe (Kraft)
21:39 Prefix codes distributions
21:46 Prefix codes distributions
22:07 General Recipe (Kraft)
23:16 Example 3: distributions codes
24:38 Example 3: distributions codes
25:57 The Most Important Slide!
26:37 Remarks
27:39 Part I: Overview
27:52 Universal Codes
29:11 Universal Codes
30:51 Universal Codes
33:04 Universal Codes
33:15 Universal Codes
37:06 Terminology
38:01 Bayesian Mixtures are universal models
38:55 Bayesian Mixtures are universal models
39:55 2-part MDL code is a universal model (code)
40:49 Bayesian Mixtures are universal models
41:36 2-part MDL code is a universal model (code)
42:03 2-part vs. Bayes universal models
42:06 Optimal Universal Model
43:27 Optimal Universal Model - II
45:04 MDL Model Selection
45:40 MDL Model Selection

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 47:38
!NOW PLAYING
Watch Part 2
Part 2 44:43
Watch Part 3
Part 3 44:57
Watch Part 4
Part 4 46:51

Description

We give a tutorial introduction to the *modern* Minimum Description Length (MDL) Principle, taking into account the many refinements and developments that have taken place in the 1990s. These do not seem to be widely known outside the information theory community. We will especially emphasize the use of MDL in classification. We also consider the connections between MDL, Bayesian inference, maximum entropy inference and structural risk minimization.

See Also:

Launch Windows Media PlayerLaunch in a standalone WM Player

WMedia PlayerSwitch to Windows Media Player

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 sally, February 13, 2008 at 5:52 p.m.:

hi!!! i'm sally.I would like to ask internet to give some nice introduction to modern watches..because the ones who has to do a project or something else on modern watches they cant find the introduction on modern watches or anything...my friend she had to do a project on modern watches and she could'nt find any of the information she has to get from internet....so i would like to do a request..please put some good information on modern watches and other items too...please please please please that is only my request.....please please please please fulfil my request...:(

Write your own review or comment:

make sure you have javascript enabled or clear this field: