RATES OF CONVERGENCE IN ACTIVE LEARNING
RATES OF CONVERGENCE IN ACTIVE LEARNING
复制标题
DOI：
10.1214/10-aos843
复制
发表时间：
2011-02-01
影响因子：
4.5
通讯作者：
Hanneke, Steve
中科院分区：
文献类型：
--
作者：
Hanneke, Steve
我们研究了在各种类型的标签噪声下，通过主动学习可实现的泛化误差的收敛速度。此外，我们研究的一般问题的模型选择与嵌套层次结构的假设类的主动学习，并提出了一种算法，其错误率可证明收敛到最佳实现的错误之间的分类器在层次结构中的速度自适应的最佳分类器的复杂性和噪声条件。特别是，我们陈述了这些速率比被动学习所能达到的速率快得多的充分条件。
We study the rates of convergence in generalization error achievable by active learning under various types of label noise. Additionally, we study the general problem of model selection for active learning with a nested hierarchy of hypothesis classes and propose an algorithm whose error rate provably converges to the best achievable error among classifiers in the hierarchy at a rate adaptive to both the complexity of the optimal classifier and the noise conditions. In particular, we state sufficient conditions for these rates to be dramatically faster than those achievable by passive learning.