Dynamic Feature Selection for Classification on a Budget
Dynamic Feature Selection for Classification on a Budget
复制标题
用于预算分类的动态特征选择
DOI：
--
复制
发表时间：
2013
期刊：
影响因子：
--
通讯作者：
Trevor Darrell
中科院分区：
文献类型：
--
作者：
Sergey Karayev;Mario Fritz;Trevor Darrell
选择的特征子集Hπ（x）的成本是CHπ（x）。通过仅接受CH ≤ B的答案，对数据敏感的损耗LB呈现硬预算约束。此外，LB可以是成本敏感的：以更低的成本给出的答案比更昂贵的答案更有价值。后一个属性的动机是Anytime性能;我们应该能够在任何时候停止算法的执行，并获得最佳答案
The cost of a selected feature subset Hπ(x) is CHπ(x). The budget-sensitive loss LB presents a hard budget constraint by only accepting answers with CH ≤ B. Additionally, LB can be cost-sensitive: answers given with less cost are more valuable than costlier answers. The motivation for the latter property is Anytime performance; we should be able to stop our algorithm’s execution at any time and have the best possible answer