图 书 馆
布莱克威尔的平易近人和无悔学习是等价的Blackwell Approachability and No-Regret Learning are Equivalent |
|
| 课程网址: | http://videolectures.net/colt2011_abernethy_learning/ |
| 主讲教师: | Jacob Abernethy |
| 开课单位: | 加州大学 |
| 开课时间: | 2011-08-02 |
| 课程语种: | 英语 |
| 中文简介: |  我们考虑了两个具有向量收益的玩家游戏的著名布莱克韦尔可接近性定理。布莱克韦尔本人此前曾证明，该定理暗示着一个简单的在线学习问题的“不后悔”算法的存在。我们证明了这种关系实际上要牢固得多，在非常强烈的意义上，布莱克韦尔的结果等同于在线线性优化的后悔最小化问题。我们表明，针对一个这样的问题的任何算法都可以有效地转换为针对另一个问题的算法。我们提供了这种减少方法的一种新颖应用：第一个有效的用于校准预测的算法。 |
| 课程简介: | We consider the celebrated Blackwell Approachability Theorem for two-player games with vector payoffs. Blackwell himself previously showed that the theorem implies the existence of a “no regret” algorithm for a simple online learning problem. We show that this relationship is in fact much stronger, that Blackwell’s result is equivalent to, in a very strong sense, the problem of regret minimization for Online Linear Optimization. We show that any algorithm for one such problem can be efficiently converted into an algorithm for the other. We provide one novel application of this reduction: the first efficient algorithm for calibrated forecasting. |
| 关 键 词: | 校准预测; 线性优化 |
| 课程来源: | 视频讲座网 |
| 数据采集: | 2020-12-29：zyk |
| 最后编审: | 2020-12-29：zyk |
| 阅读次数: | 105 |