图 书 馆
经验Bernstein停止Empirical Bernstein Stopping |
|
| 课程网址: | http://videolectures.net/icml08_mnih_ebs/ |
| 主讲教师: | Volodymyr Mnih |
| 开课单位: | 多伦多大学 |
| 开课时间: | 2008-08-29 |
| 课程语种: | 英语 |
| 中文简介: |  |
| 课程简介: | Sampling is a popular way of scaling up machine learning algorithms to large datasets. The question often is how many samples are needed. Adaptive stopping algorithms monitor the performance in an online fashion and make it possible to stop early, sparing valuable computation time. We concentrate on the setting where probabilistic guarantees are desired and demonstrate how recently-introduced empirical Bernstein bounds can be used to design stopping rules that are efficient. We provide upper bounds on the sample complexity of the new rules as well as empirical results on model selection and boosting in the filtering setting. |
| 关 键 词: | 机器学习; 监控性能; 经验结果 |
| 课程来源: | 视频讲座网 |
| 数据采集: | 2023-03-14：chenjy |
| 最后编审: | 2023-03-14：chenjy |
| 阅读次数: | 15 |