0.25
0.5
0.75
1.25
1.5
1.75
2
The Graph-guided Group Lasso
Published on Aug 26, 20134044 Views
In this work we propose a penalised regression model in which the covariates are known to be clustered into groups, and the clusters are arranged as nodes in a graph. We are motivated by an applicatio
Related categories
Chapter list
The Graph-guided Group Lasso00:00
Outline - 100:03
Outline - 200:31
Outline - 300:49
Outline - 400:51
A 30s introduction to the biology00:58
Single-nucleotide polymorphisms (SNPs)01:12
Genome-wide association study (GWAs)01:29
Notation01:41
Sparse solution02:08
Penalized linear regression - 102:30
Penalized linear regression - 202:38
Some notable penalties that impose sparsity02:52
Incorporating prior biological knowledge - Variable grouping - 102:55
Incorporating prior biological knowledge - Variable grouping - 203:11
Incorporating prior biological knowledge - Variable grouping - 303:17
Incorporating prior biological knowledge - Variable grouping - 403:26
Incorporating prior biological knowledge - Variable grouping - 503:44
Incorporating prior biological knowledge - Network - 104:00
Incorporating prior biological knowledge - Network - 204:18
Incorporating prior biological knowledge - Network - 304:36
Incorporating prior biological knowledge - Network - 404:42
Incorporating prior biological knowledge - Network - 504:47
Incorporating prior knowledge at multiple levels05:36
The between-group relations06:12
Notation06:41
GGGL-1: Illustration07:30
GGGL-1: The model07:59
GGGL-1: Smoothing e ffect09:18
GGGL-1: A potential side e ffect10:35
GGGL-2: Another interpretation11:45
GGGL-2: The model - 112:10
GGGL-2: The model - 212:29
GGGL-2: Smoothing e ffect12:49
GGGL-2: Within-group eff ect13:27
Comparison: GGGL-1 and GGGL-2 smoothing e ffect - 113:46
Comparison: GGGL-1 and GGGL-2 smoothing e ffect - 213:56
Data generation: key settings - 114:12
Data generation: key settings - 214:34
Comparison: small for GGGL-115:56
Comparison: large for GGGL-116:33
Comparison: small for GGGL-217:16
Comparison: large for GGGL-217:54
Estimation algorithm: GGGL-1 - 118:07
Estimation algorithm: GGGL-1 - 218:21
Estimation algorithm: GGGL-1 - 318:27
Estimation algorithm: GGGL-1 - 418:37
Estimation algorithm: GGGL-1 - 518:44
Estimation algorithm: GGGL-1 - 619:04
Estimation algorithm: GGGL-2 - 119:18
Estimation algorithm: GGGL-2 - 219:31
Parallel computation: outline - 119:44
Parallel computation: outline - 220:27
Preliminary results21:53
Networks for GGGL22:26
Illustration of networks23:50
Experiment design: GGGL-1 vs Group lasso24:48
GGGL-1 vs Group lasso24:52
Future works25:08
Acknowledgement26:13
Reference26:30