en-de
en-es
en-fr
en-sl
en
en-zh
0.25
0.5
0.75
1.25
1.5
1.75
2
Kernel-based learning of hierarchial multilabel classification models
Published on 2007-02-253292 Views
Related categories
Presentation
Kernel-based learning of hierarchial00:01
Hierarchical Multilabel Classification:00:53
How to learn hierarchical multilabels?02:12
How to measure loss?02:36
Scaling the loss06:22
The classification model09:18
Feature vectors11:19
From Maximum Likelihood to Maximum Margin13:43
Scaling the margin15:55
Primal optimization problem17:41
Dual problem18:51
Solving the optimization problem19:53
Marginalizing the problem21:04
Ensuring marginal consistency22:50
Marginalized problem23:55
Decomposing the problem24:59
The optimization algorithm26:46
Conditional Gradient Ascent30:51
Working set maintenance31:27
Experiments32:54
Example learning curve34:44
Results36:27
Conclusions38:18