en
0.25
0.5
0.75
1.25
1.5
1.75
2
Attributes for classifier feedback
Published on Nov 12, 20126407 Views
Traditional active learning allows a (machine) learner to query the (human) teacher for labels on examples it finds confusing. The teacher then provides a label for only that instance. This is quite
Related categories
Chapter list
Attributes for Classifier Feedback00:00
You can teach a child by examples...00:20
Giraffe00:29
Elephant00:32
Pink Cow00:34
Giraffe00:36
Sheep00:39
Giraffe00:41
Cat00:43
And on, and on, and on…00:46
Bored child00:51
Enthusiastic child01:10
Learning01:15
And on, and on, and on ...01:42
Child01:46
Proposed Active Learning Scenario01:56
Current belief (1)02:06
Current belief (2)03:15
Communication03:34
Proposed Active Learning04:04
Relative Attributes05:36
Attributes-based Feedback (1)07:00
Attributes-based Feedback (2)07:07
Proposed Active Learning07:40
Label-based Feedback (1)07:46
Label-based Feedback (2)08:27
Label-based Feedback (3)09:09
Datasets (1)10:24
Datasets (2)10:29
Datasets (3)10:36
Settings10:47
Results (1)11:13
Results (2)12:07
Conclusion12:42
Thank you!13:20