On Multilabel Classification and Ranking with Partial Feedback
Abstract
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 exploration and exploitation. We analyze this algorithm in a partial adversarial setting, where covariates can be adversarial, but multilabel probabilities are ruled by (generalized) linear models. We show $O(T^{1/2}\log T)$ regret bounds, which improve in several ways on the existing results. We test the effectiveness of our upper-confidence scheme by contrasting against full-information baselines on real-world multilabel datasets, often obtaining comparable performance.
Cite
Text
Gentile and Orabona. "On Multilabel Classification and Ranking with Partial Feedback." Neural Information Processing Systems, 2012.Markdown
[Gentile and Orabona. "On Multilabel Classification and Ranking with Partial Feedback." Neural Information Processing Systems, 2012.](https://mlanthology.org/neurips/2012/gentile2012neurips-multilabel/)BibTeX
@inproceedings{gentile2012neurips-multilabel,
title = {{On Multilabel Classification and Ranking with Partial Feedback}},
author = {Gentile, Claudio and Orabona, Francesco},
booktitle = {Neural Information Processing Systems},
year = {2012},
pages = {1151-1159},
url = {https://mlanthology.org/neurips/2012/gentile2012neurips-multilabel/}
}