Multi-task Learning  thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Multi-task Learning

Published on Aug 26, 20138044 Views

A fundamental limitation of standard machine learning methods is the cost incurred by the preparation of the large training samples required for good generalization. A potential remedy is o ffered by

Related categories

Chapter list

Multi-task Learning00:00
Outline00:13
Problem Formulation (cont.)07:01
Applications07:09
Problem Formulation07:13
Examples of Regularizers09:41
Learning Sparse Representations13:36
Connection to Sparse Coding22:19
Experiments22:20
Experiments (cont.)22:34
Learning Bound27:20
Analysis of Learning to Learn35:22
Comparison to Sparse Coding Bound36:41
Multilinear MTL38:04
Multilinear MTL (cont.) - 243:54
Multilinear MTL (cont.) - 147:12
Alternative Convex Relaxation47:22
Quality of Relaxation (cont.)48:29
Problem Reformulation49:16
ADMM49:22
Proximity Operator49:24
Experiments49:31
Conclusions50:19