图 书 馆
细粒度的分类和部分反馈的排名On Multilabel Classification and Ranking with Partial Feedback |
|
| 课程网址: | http://videolectures.net/machine_orabona_ranking/ |
| 主讲教师: | Francesco Orabona |
| 开课单位: | 芝加哥丰田技术学院 |
| 开课时间: | 2013-01-14 |
| 课程语种: | 英语 |
| 中文简介: |  |
| 课程简介: | 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(T1/2logT) 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. |
| 关 键 词: | 局部信息设置工作; 细粒度的概率; 线性模型 |
| 课程来源: | 视频讲座网 |
| 最后编审: | 2020-06-29：zyk |
| 阅读次数: | 44 |