0.25
0.5
0.75
1.25
1.5
1.75
2
Output Kernel Learning Methods
Published on Aug 26, 20134417 Views
A rather flexible approach to multi-task learning consists in solving a regularization problem where a suitable kernel is used to model joint relationships between both inputs and tasks. Since specif
Related categories
Chapter list
Output Kernel Learning Methods00:00
Part 100:21
Multiple regression (1)00:22
Multiple regression (2)01:44
Multiple regression (3)02:20
Collaborative filtering (1)02:23
Collaborative filtering (2)03:50
Multi-task learning (1)04:41
Multi-task learning (2)04:57
Multi-task learning (3)05:14
Object recognition (1)05:27
Graph (1)06:02
Part 206:32
Kernel-based multi-task learning 06:42
Decomposable kernels08:36
Kernel-based regularization methods09:31
Multiple Kernel learning12:17
Output Kernel learning13:56
Low-Rank Output Kernel learning15:41
Part 318:35
MovieLens datasets18:39
Results20:31
Object recognition (2)21:48
Graph (2)22:24
Conclusions22:42
References23:26