en
en-es
en-fr
en
en
en-sl
en
0.25
0.5
0.75
1.25
1.5
1.75
2
Efficient Active Learning
Published on Feb 4, 20254558 Views
We present and analyze an active learning algorithm that is theoretically sound in an agnostic setting, empirically effective, and as efficient as standard online learning algorithms. This allows us t
Related categories
Presentation
Efficient Active Learning00:00
Active Learning00:09
Exploration vs Exploitation01:42
Importance Weighted Active Learning02:07
How should p be chosen? (1)04:31
How should p be chosen? (2)05:11
How should p be chosen? (3)05:22
How should p be chosen? (4)05:53
How should p be chosen? (5)06:17
Theorems06:32
Active Learning with Online Gradient (1)07:14
Active Learning with Online Gradient (2)07:49
Active Learning with Online Gradient (3)08:03
Active Learning with Online Gradient (4)08:21
Active Learning with Online Gradient (5)08:24
Learning with importance weights (1)09:06
Learning with importance weights (2)09:29
Learning with importance weights (3)10:40
Learning with importance weights (4)11:01
Online Gradient Descent Results11:22
How fast is it? (1)12:53
How fast is it? (2)13:49
Conclusions14:31