A Tensor Spectral Approach to Learning Mixed Membership Community Models thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

A Tensor Spectral Approach to Learning Mixed Membership Community Models

Published on Aug 09, 20134108 Views

Modeling community formation and detecting hidden communities in networks is a well studied problem. However, theoretical analysis of community detection has been mostly limited to models with non-ove

Related categories

Chapter list

A Tensor Approach to LearningMixed Membership Community Models00:00
Community00:11
Stochastic Block Model00:58
Overlapping Communities? - 101:48
Overlapping Communities? - 202:21
Overlapping Communities? - 302:29
Mixed Membership Stochastic Block (MMSB) Model - 102:49
Mixed Membership Stochastic Block (MMSB) Model - 203:19
Mixed Membership Stochastic Block (MMSB) Model - 304:13
Mixed Membership Stochastic Block (MMSB) Model - 404:37
MMSB Model - 104:46
MMSB Model - 204:58
MMSB Model - 305:17
MMSB Model - 405:23
MMSB Model - 505:36
MMSB Model - 605:40
MMSB Model - 705:53
Our Results - 207:28
Our Results - 107:48
Method of moments: Graph Moments - 108:12
Method of moments: Graph Moments - 208:46
Method of moments: Graph Moments - 308:55
Method of moments: Graph Moments - 409:19
Method of moments: Graph Moments - 509:51
Why 3-star?10:20
Analog to Topic Modeling - 111:09
Analog to Topic Modeling - 211:27
Analog to Topic Modeling - 311:43
Analog to Topic Modeling - 411:44
Analog to Topic Modeling - 511:51
Analog to Topic Modeling - 612:02
Analog to Topic Modeling - 712:25
Difficulty: we are not docs and words13:17
Algorithm Outline14:21
Simultaneous Whitening - 115:16
Simultaneous Whitening - 215:48
Tensor Power Method[AGHKT 12]16:04
Thresholding16:48
Open Problems17:51
Thank you18:36