0.25
0.5
0.75
1.25
1.5
1.75
2
Active and passive learning of linear separators under log-concave distributions
Published on Aug 09, 20137750 Views
We prove that active learning provides an exponential improvement over PAC (passive) learning of homogeneous linear separators under nearly log-concave distributions. Building on this, we provide a co
Related categories
Chapter list
Active and passive learning of linear separators00:00
2-Minute Version - 100:07
2-Minute Version - 201:37
Supervised Learning Formalization02:21
Sample Complexity Results03:11
Active Learning05:05
Classic Paradigm Insufficient Nowadays05:42
When Active Learning Helps05:54
Margin Based Active Learning - 107:09
Margin Based Active Learning - 208:24
Margin Based Active-Learning, Realizable Case - 108:47
Margin Based Active-Learning, Realizable Case - 209:28
Margin Based Active-Learning, Realizable Case - 310:23
Linear Separators, Log-Concave Distributions - 111:06
Linear Separators, Log-Concave Distributions - 212:31
Linear Separators, Log-Concave Distributions - 313:48
Margin Based Active-Learning, Realizable Case - 414:00
Proof Idea - 115:20
Proof Idea - 215:35
Passive Learning - 116:45
Passive Learning - 218:41
Discussion, Open Directions21:17