0.25
0.5
0.75
1.25
1.5
1.75
2
Semi-supervised tree learning
Published on Jan 31, 20171451 Views
Related categories
Chapter list
Semi-supervised tree learning for SOP00:00
Outline - 100:53
Motivation - 101:16
Motivation - 201:57
SSL for classification tasks02:48
SSL for regression tasks03:02
SSL for multi-label classification03:10
SSL for multi-target regression03:19
Motivation03:30
SSL underlying mechanism05:00
Existing approaches05:30
Existing approaches: Issues06:13
Maestra: Extend PCT framework towards SSL07:40
Outline - 208:08
Predictive clustering08:24
Predictive clustering trees - 109:23
Predictive clustering trees - 209:33
PCTs instantiations09:47
Outline - 310:42
Supervised PCTs - 110:54
Supervised PCTs - 211:25
Supervised PCTs - 311:49
Semi-supervised PCTs12:06
Semi-supervised PCTs: mixed attributes - 113:09
Semi-supervised PCTs: mixed attributes - 213:32
Semi-supervised PCTs for SOP13:54
SSL PCTs: Smoothness in the target space - 114:11
SSL PCTs: Smoothness in the target space - 216:02
SSL PCTs extensions17:11
Semi-supervised PCTs for MTR - 117:38
Semi-supervised PCTs for MTR - 218:07
How we handle extreme cases?18:48
Feature weighted semi-supervised PCTs19:49
Experimental design20:27
Results: Statistical analysis - 122:33
Results: Statistical analysis - 223:19
Results: Statistical analysis - 323:37
Results: Statistical analysis - 423:44
Results: Per-dataset - 124:17
Results: Per-dataset - 224:51
Results: Per-dataset - 325:02
Results: Influence of the w parameter - 125:12
Results: Influence of the w parameter - 225:52
Results: Influence of the w parameter - 326:42
Results: Influence of the w parameter - 427:10
Results: Variable amount of unlabeled data27:33
Example PCTs28:28
Example semi-supervised PCT29:12
Summary29:35
SSL PCTs for classification tasks30:32
Binary classification results31:27
Multi-class classification results34:10
Binary classification: SSL PCTs example34:38
MC classification: SSL PCTs example35:07
Binary/multi-class classification summary35:37
SSL PCTs for multi-label classification36:32
Sample results: multi-label classification37:13
Conclusions37:54
Further work38:49