en
0.25
0.5
0.75
1.25
1.5
1.75
2
A boosting approach to multiview classification with cooperation
Published on 2011-10-033371 Views
Nowadays in numerous elds such as bioinformatics or multimedia, data may be described using many di erent sets of features (or views) which carry either global or local information. Many learning t
Related categories
Presentation
A boosting approach to multiview classi cation with cooperation00:00
Outline - 100:17
Outline - 200:40
Multi view learning for image classi cation - 100:43
Multi view learning for image classi cation - 201:41
Multi view learning : Fusion based methods01:49
Motivation02:35
Weak views04:13
Multiclass boosting - 104:32
Multiclass boosting - 205:00
Multiclass boosting - 305:15
Outline - 305:52
The core of MuMBo05:57
Schema of the algorithm07:11
Schema of the algorithm : matrix update formula08:22
A bound per view09:34
Proof of the bound - 110:02
Generalization error bound11:19
Proof of the bound - 212:01
Outline - 412:32
Results on synthetic data12:34
Outline - 514:00
Conclusion14:01
Questions15:22