On Multilabel Classification and Ranking with Partial Feedback thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

On Multilabel Classification and Ranking with Partial Feedback

Published on Jan 14, 20132670 Views

We present a novel multilabel/ranking algorithm working in partial information settings. The algorithm is based on 2nd-order descent methods, and relies on upper-confidence bounds to trade-off explora

Related categories

Chapter list

On Multilabel Classification and Ranking with Partial Feedback00:00
Cost sensitive, hamming, and ranking loss00:51
The probability to click on a banner is given by a Generalized Linear Model02:02
Experiments: performance comparable to full-information02:56