Learning Theory

author:Mark Reid, Research School of Information Sciences and Engineering, Australian National University
published: April 1, 2009, recorded: January 2009, views: 527

Slides

Slides
0:00	Information, Divergence and Risk for Binary Classification
1:02	Taylor & Jensen’s Most Excellent Adventure
1:29	Introduction
1:43	The Blind Men & The Elephant (1)
2:37	The Blind Men & The Elephant (2)
2:54	Jules Henri Poincaré
3:11	What’s in it for me? (1)
3:32	What’s in it for me? (2)
3:42	What’s in it for me? (3)
3:54	Terra Statistica (1)
4:14	Terra Statistica (2)
4:26	Terra Statistica (3)
4:44	Terra Statistica (4)
4:54	Terra Statistica (5)
5:06	Part I: Convexity, Binary Experiments & Classification
5:10	Convexity
5:13	Convex Sets (1)
6:09	Convex Sets (2)
6:53	Convex Sets (3)
7:07	Convex Functions
7:37	Taylor’s Theorem (1)
8:54	Taylor’s Theorem (2)
10:39	Integral Form of the Taylor Expansion
11:50	Bregman Divergence (1)
12:37	Bregman Divergence (2)
13:09	Jensen’s Inequality (1)
14:05	Jensen’s Inequality (2)
14:20	Jensen’s Inequality (3)
14:50	Jensen’s Inequality
15:05	The Legendre-Fenchel Transform (1)
16:26	The Legendre-Fenchel Transform (2)
16:55	Representations of Convex Functions
18:07	Terra Statistica (1)
18:16	Terra Statistica (2)
18:19	Binary Experiments and Measures of Divergence
18:22	Binary Experiments (1)
19:13	Binary Experiments (2)
19:58	Binary Experiments (3)
20:35	Test Statistics (1)
20:43	Test Statistics (2)
21:24	Test Statistics (3)
22:08	Statistical Power and Size
23:18	The Neyman-Pearson Lemma (1)
24:16	The Neyman-Pearson Lemma (2)
26:35	Csiszár f-Divergence (1)
28:38	Csiszár f-Divergence (2)
29:01	Csiszár f-Divergence (3)
29:22	Csiszár f-Divergence
30:06	Examples (1)
32:47	Examples (2)
33:28	Examples (3)
34:41	Examples (4)
35:05	Terra Statistica (1)
35:09	Terra Statistica (2)
35:51	Classification and Probability Estimation
35:54	From Hypothesis Testing to Classification (1)
36:30	From Hypothesis Testing to Classification (2)
37:22	Generative and Discriminative Views (1)
38:53	Generative and Discriminative Views (2)
39:38	Generative and Discriminative Views (3)
40:56	Loss, Risk and Regret (1)
42:40	Loss, Risk and Regret (2)
43:31	Loss, Risk and Regret (3)
43:53	Loss, Risk and Regret (4)
44:35	Loss, Risk and Regret (5)
45:18	Loss Examples
46:46	Fisher Consistency & Proper Losses (1)
48:08	Fisher Consistency & Proper Losses (2)
48:53	Examples of Proper Losses
51:07	Properties of Proper Losses (1)
52:13	Properties of Proper Losses (2)
53:11	Savage’s Theorem
55:41	Examples (1)
55:50	Examples (2)
56:05	Examples (3)
56:20	Examples (4)
56:43	- Savage’s Theorem-Part 2
56:55	Examples (5)
57:10	Examples (6)
57:16	Examples (7)
57:42	Proper Point-wise Bayes Risks
58:52	Information
59:01	T.S. Eliot
59:27	Statistical Information (1)
60:18	Statistical Information (2)
60:56	Statistical Information (3)
61:45	Statistical Information (4)
62:59	Statistical Information (5)
63:27	Statistical Information (6)
63:49	Statistical Information
64:09	Bregman Information
65:24	Jules Henri Poincaré

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.

Watch videos: (click on thumbnail to launch)

Part 1 1:05:47 Flash video

!NOW PLAYING

Part 2 0:54:11 Flash video

Part 3 0:44:23 Flash video

Description

This course highlights some relationships between surrogate losses, scoring rules, f-divergences, Bregman divergences, statistical information and ROC curves and their implications for applications such as divergence estimation.