en-de
en-es
en-fr
en-pt
en-sl
en
en-zh
0.25
0.5
0.75
1.25
1.5
1.75
2
Neyman-Pearson classification under a strict constraint
Published on Aug 02, 20115022 Views
Motivated by problems of anomaly detection, this paper implements the Neyman-Pearson paradigm to deal with asymmetric errors in binary classification with a convex loss. Given a finite collection of c
Related categories
Chapter list
Neyman-Pearson classification00:00
Binary classification00:00
Neyman-Pearson paradigm00:43
Asymmetry in the errors02:08
Some examples of binary classification from FHT - 102:40
Some examples of binary classification from FHT - 202:58
Some examples of binary classification from FHT - 303:09
Some examples of binary classification from FHT - 403:17
Some examples of binary classification from FHT - 503:29
Some examples of binary classification from FHT - 603:33
Some examples of binary classification from FHT - 703:42
Some examples of binary classification from FHT - 803:47
Classification error may be dangerous04:09
Observations05:03
Previous work06:36
Idea of Cannon et al.: relaxed constraint - 107:32
Idea of Cannon et al.: relaxed constraint - 208:15
Comments09:30
Empirical risk10:06
Contribution of this talk11:06
Convexification11:46
Convexified NP problem13:05
Neyman-Pearson classifier14:23
Theorem - 114:49
An assumption15:27
Theorem - 216:26
Sketch of the proof17:14
Some intuition - 117:57
Some intuition - 218:11
Some intuition - 318:20
Proposition18:27
Chance constrained optimization - 119:14
Chance constrained optimization - 219:32
Chance constrained optimization - 320:04