High-Dimensional Graphical Model Selection thumbnail
slide-image
Pause
Mute
Subtitles not available
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

High-Dimensional Graphical Model Selection

Published on Jan 25, 20128312 Views

We consider the problem of Ising and Gaussian graphical model selection given n i.i.d. samples from the model. We propose an efficient threshold-based algorithm for structure estimation based known as

Related categories

Chapter list

High-Dimensional Graphical Model Selection00:00
Graphical Models: Definition - 100:23
Graphical Models: Definition - 200:51
Graphical Models: Definition - 301:07
Graphical Models: Definition - 401:08
Structure Learning of Graphical Models - 101:17
Structure Learning of Graphical Models - 201:43
Tree Graphical Models: Tractable Learning - 102:01
Tree Graphical Models: Tractable Learning - 202:12
Tree Graphical Models: Tractable Learning - 302:29
Tree Graphical Models: Tractable Learning - 402:29
Tree Graphical Models: Tractable Learning - 502:53
Learning Graphical Models Beyond Trees - 103:11
Learning Graphical Models Beyond Trees - 203:36
Learning Graphical Models Beyond Trees - 304:04
Learning Graphical Models Beyond Trees - 404:27
Related Work in Structure Learning04:45
Outline - 105:31
Intuitions: Conditional Mutual Information Test - 105:32
Intuitions: Conditional Mutual Information Test - 206:00
Intuitions: Conditional Mutual Information Test - 306:42
Intuitions: Conditional Mutual Information Test - 406:49
Intuitions: Conditional Mutual Information Test - 506:55
Tractable Graph Families: Local Separation07:12
Outline - 210:25
Setup: Ising and Gaussian Graphical Models - 110:26
Setup: Ising and Gaussian Graphical Models - 210:28
Setup: Ising and Gaussian Graphical Models - 310:29
Setup: Ising and Gaussian Graphical Models - 410:29
Setup: Ising and Gaussian Graphical Models - 510:30
Regime of Tractable Learning - 111:01
Regime of Tractable Learning - 211:02
Regime of Tractable Learning - 312:03
Tractable Graph Families and Regimes - 112:26
Tractable Graph Families and Regimes - 212:27
Tractable Graph Families and Regimes - 312:27
Tractable Graph Families and Regimes - 412:28
Example: girth g, maximum degree Δ13:11
Outline - 314:08
Algorithm for Structure Learning - 114:09
Algorithm for Structure Learning - 214:10
Algorithm for Structure Learning - 314:59
Guarantees on Conditional Mutual Information Test15:02
Lower Bound on Sample Complexity - 115:31
Lower Bound on Sample Complexity - 215:33
Proof Ideas - 115:41
Proof Ideas - 215:42
Proof Ideas - 315:43
Proof Ideas - 416:12
Outline - 416:13
Summary and Outlook16:14