A Close Look to Margin Complexity and Related Parameters thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

A Close Look to Margin Complexity and Related Parameters

Published on Aug 02, 20113077 Views

Concept classes can canonically be represented by sign-matrices, i.e., by matrices with entries 1 and −1. The question whether a sign-matrix (concept class) A can be learned by a machine that perfor

Related categories

Chapter list

A Close Look to Margin Complexity and Related Parameters00:00
Overview - 100:19
Large margin classification - 100:28
Large margin classification - 200:53
Large margin classification - 301:10
Margin parameters - 101:28
Margin parameters - 202:20
Margin parameters - 302:55
Margin parameters - 403:35
Margin parameters - 504:21
Our Contribution - 104:39
Our Contribution - 205:06
Our Contribution - 305:48
Overview - 206:22
Margin maximization and its dual - 106:28
Margin maximization and its dual - 206:59
Proof sketch - 107:40
Proof sketch - 208:31
Proof sketch - 309:05
Proof sketch - 409:17
"Support Vectors" for Linear Arrangements - 109:58
"Support Vectors" for Linear Arrangements - 210:30
"Support Vectors" for Linear Arrangements - 310:55
Overview - 311:53
Connection to CSQdim - 112:00
Connection to CSQdim - 212:19
Connection to CSQdim - 312:51
Connection to CSQdim - 413:15
Connection to CSQdim - 514:07
Connection to CSQdim - 614:59
Connection to CSQdim - 715:06
Connection to CSQdim - 815:36
Connection to CSQdim - 916:03
Connection to CSQdim - 1016:19
Connection to CSQdim - 1116:32
Overview - 416:43
Summing up - 116:49
Summing up - 217:10
Summing up - 317:25
Open problems18:51
Thank you19:41